NJR 18th Annual Report 2021

HIPS
KNEES
ANKLES
ELBOWS
SHOULDERS
National Joint Registry
18th Annual Report
ISSN 2054-183X (Online)

2021
Surgical data to 31 December 2020
Prepared by
NJR Editorial Board and contributors
NJRSC Members
Mike Reed (Chairman, Editorial Board)
Robin Brittain
Peter Howard
Sandra Lawrence
Jeffrey Stonadge
Mark Wilkinson
Timothy Wilton
NJR RCC Representatives

Derek Pegg (Chairman, RCC Committee)
Sebastian Dawson-Bowling
Adam Watts
Orthopaedic Specialists
Colin Esler
Andy Goldberg
Simon Jameson
Toby Jennison
Andrew Toms
NJR Management Team

Elaine Young
Chris Boulton
Deirdra Taylor
Oscar Espinoza
NEC Software Solutions UK Ltd

NJR data management, data solutions and associated services
Victoria McCormack
Claire Newell
Martin Royall
Mike Swanson
University of Bristol / University of Oxford

NJR statistical analysis, support and associated services
Yoav Ben-Shlomo
Ashley Blom
Emma Clark
Kevin Deere
Celia Gregson
Andrew Judge
Erik Lenguerrand
Andrew Price
Dani Prieto-Alhambra
Jonathan Rees
Adrian Sayers
Michael Whitehouse
Pad Creative Ltd (design and production)
Additional data and information can also be found as outlined on pages 4-6.
National Joint Registry | 18th Annual Report
Introduction
The National Joint Registry (NJR) collects information results are also shared with implant manufacturers.
about hip, knee, ankle, elbow and shoulder joint The report also includes some short excerpts which
replacement operations (arthroplasty) from all showcase the NJR’s contribution to orthopaedic
participating hospitals in England, Wales, Northern research activity, illustrating the value of the use of this
Ireland, the Isle of Man and the States of Guernsey. As collected data.
the largest data collection of its kind in the world, the
NJR has been described in UK Parliament as a global The work of the NJR and the
exemplar of an implantable medical devices registry. contribution of patients
The registry’s purpose is to record patient information The registry has shown that orthopaedic surgery, as
and provide data on: the performance and longevity of one of the main uses of implant devices in the UK, is
replacement joint implants; the surgical outcomes for demonstrating the highest standards of patient safety
the hospitals where these operations are carried out; with regard to their use. Patient representatives are
and on the performance outcomes of the surgeons actively involved in our workstreams and committees.
who conduct the procedures. With well over three million records, registry data are
also made available under strict security conditions to
We produce this Annual Report, summarising our
medical and academic researchers, to further progress
work and sharing the analysis of data for the past year,
the pool of work in measuring and understanding which
visually in tables and graphs, for procedures across
practices provide better outcomes.
each of the joints, as well as implant and hospital
outcomes. The report illustrates how the number of Our data collection and analysis work provides the
elective joint procedures has been heavily impacted evidence to drive continuous development and
throughout this past year by COVID-19; many joint implementation of measures, to ensure implant safety
procedures have had to be cancelled or postponed. and the enhancement of patient outcomes is always
There has been a major impact on people awaiting top of the agenda alongside a focus on reduced
surgery for all joint types in a climate of continued revision rates year on year; as well as improvements
uncertainty for their future surgical dates, with in standards in quality of care, whilst also addressing
increasingly extended waiting lists. We have covered overall cost-effectiveness in joint replacement surgery.
this impact, both on the orthopaedic sector and with
a patient perspective on those who are facing long We are very grateful to all patients, who having
waiting times, in a special feature in this year’s report, undergone a joint replacement, have provided their
you can find this on page 341. data to the NJR over the years, which has enabled us
to collect and develop such a rich and valuable data
Registry data for the surgery that has taken place source. The registry is also appreciative of the work
this past year have again been analysed by expert of data entry staff in all participating hospitals, who
statisticians and the results published with the willingly engage in our stringent data quality award
continued aim of enhancing safety and improving programmes to ensure our information is of high
clinical outcomes for the benefit of patients and the quality, accurate and as complete as is possible.
whole orthopaedic healthcare sector - device outcome
This work uses data provided by patients and collected by

hospitals as part of their care and support.
www.njrcentre.org.uk 3
Summary of content for the NJR Annual Report
Summary Content Full information can be found
Introduction to the NJR and Foreword from the In this report and via
Introduction
NJR Steering Committee Chairman reports.njrcentre.org.uk
Summary of this year’s report by the NJR

In this report and via
Executive summary Editorial Board Chairman and NJR Medical
reports.njrcentre.org.uk
Director
Statistics on joint replacement activity for hip,

reports.njrcentre.org.uk through
Clinical activity 2020 knee, ankle, elbow and shoulder activity for the
interactive reporting
period 1 January to 31 December 2020
Detailed statistical analyses on hip and knee

replacement surgery using data from 1 April 2003
to 31 December 2020. Analysis of primary ankles
Outcomes after joint replacement
and shoulders representing data collected since In this report
surgery 2003-2020
1 January 2010 and 1 April 2012 respectively.
Analyses on data for elbows using data collected
since 1 April 2012
Indicators for hip and knee joint replacement

procedures by Trust, Local Health Board and In this report and via
Implant and unit-level activity
unit. Plus commentary on implant performance reports.njrcentre.org.uk and
and outcomes
and those that have higher than expected rates download area
of revision and were reported to the MHRA
Information on the work of the NJR committees

Developments reports.njrcentre.org.uk
and NJR development to 31 March 2021
Composition, attendance, declarations of

NJR’s governance and reports.njrcentre.org.uk and
interest for the NJR Steering Committee, sub-
operational structure download area
committees and terms of reference
In this report and via

Published and approved research papers using
Research reports.njrcentre.org.uk and
NJR data
download area
4 www.njrcentre.org.uk
NJR Reports online

Clinical activity 2020 overview
The interactive portion of our 18th Annual Report • Procedure details by type of provider
can be found online via the registry’s dedicated • Primary procedure details by type of provider
NJR Reports website at: reports.njrcentre.org.uk.
• Types of primary replacements undertaken
Here we present data on clinical activity during the • Patient characteristics for primary replacement
2020 calendar year. This includes information on procedures, according to procedure type
the volumes and surgical techniques in relation to • Age and gender for primary replacement patients
procedures submitted to the registry, with the most
recent data being for the period 1 January 2020 to 31 • Patients’ physical status classification (ASA grades)
December 2020. To be included in these tables and for primary replacement procedures
graphs, all procedures must have been entered into • Body Mass Index (BMI) for primary
the registry by 28 February 2021. replacement patients
• Indications for primary procedure based on
This year’s printed report includes a paper illustrating
age groups
how the volume of joint procedures has been
heavily impacted by COVID-19, along with a patient • Surgical technique for primary replacement patients
perspective on those who are now facing even • Thromboprophylaxis regime for primary replacement
longer waiting times. Supplementary data analyses patients, prescribed at time of operation
information for this work can be found online at • Reported untoward intra-operative events for
reports.njrcentre.org.uk/COVID19. primary replacement patients, according to
procedure type
The double page infographic spread at the end of this
report offers a visual summary of key facts relating to • Patient characteristics for revision procedures,
the analysis of clinical activity during the 2020 calendar according to procedure type
year. This can also be downloaded as a waiting room • Indication for surgery for revision procedures
poster via reports.njrcentre.org.uk/downloads.
• Trends in use of the most commonly used brands
The information found online now includes historical
data, going back to 2005 in most cases. Using the
For hips specifically
dedicated website, readers are able to use interactive, • Components removed during hip
filterable graphs to identify the key information and revision procedures
trends associated with the following reports for hip, • Components used during single-stage hip
knee, ankle, elbow and shoulder data (where sufficient revision procedures
data are available):
• Trends in femoral head size and hip articulation
• Total number of hospitals and treatment
centres in England (including the Isle of Man and the For knees specifically
States of Guernsey), Wales and Northern Ireland • Implant constraint for primary procedures
• Number of participating hospitals and the number • Bearing type for primary procedures
and type of procedures performed
• Number of procedures undertaken as a proportion
of all procedures submitted annually
Navigating the NJR Reports online facility
What can you find at NJR Reports online?
Simply navigate the left hand tabs to view information on the volumes and surgical techniques in
relation to procedures submitted to the registry.
Top tabs: If you require

information about
specific procedures, go
Left hand tabs: Here, the straight to the data by
information is segregated clicking on the joint type
by report and information most relevant to you.
type. A wealth of updates
are available, from further
information on data
collection and quality, to
the work of our committees
There is also implant
and progress of NJR
and hospital specific
developments.
information available,
a glossary and
a downloadable
infographic to make
all the information as
Full NJR Reports website at: accessible as possible
reports.njrcentre.org.uk to all of our visitors.
Contents
Introduction 3
Summary of content for the NJR Annual Report 4
NJR Reports online 5
Clinical activity 2020 overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
Navigating the NJR Reports online facility . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
Index
1. Chairman’s Foreword 23
2. Executive Summary 28
3. Outcomes after joint replacement 2003 to 2020 35
3.1 Summary of data sources, linkage and methodology 35
Information governance and patient confidentiality . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36
Data quality . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36
Missing data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38
Patient level data linkage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39
Linkage between primaries and any associated revisions (the ‘linked files’) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40
Analytical methods and terminology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40
3.2 Outcomes after hip replacement 44
3.2.1 Overview of primary hip replacement surgery . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45
3.2.2 First revisions after primary hip surgery . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62
3.2.3 Revisions after primary hip replacement: effect of head size for selected bearing surfaces /
fixation sub-groups . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79
3.2.4 Revisions after primary hip surgery for the main stem / cup brand combinations . . . . . . . . . . . . . . . 92
3.2.5 Revisions for different causes after primary hip replacement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 100
3.2.6 Mortality after primary hip replacement surgery . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 109
3.2.7 Primary hip replacement for fractured neck of femur compared with other reasons for implantation . . 110
3.2.8 Overview of hip revision procedures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 116
3.2.9 Rates of hip re-revision . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 119
3.2.10 Reasons for hip re-revision . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 129
3.2.11 90-day mortality after hip revision . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 131
3.2.12 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 132
3.3 Outcomes after knee replacement 135
3.3.1 Overview of primary knee replacement surgery . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 136
3.3.2 First revision after primary knee surgery . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 148
3.3.3 Revisions after primary knee replacement surgery by main brands for TKR and UKR . . . . . . . . . . . 166
3.3.4 Revisions for different indications after primary knee replacement . . . . . . . . . . . . . . . . . . . . . . . . . 186
3.3.5 Mortality after primary knee surgery . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 191
3.3.6 Overview of knee revisions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 193
3.3.7 Rates of knee re-revision . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 196
3.3.8 Reason for knee re-revision . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 208
3.3.9 90-day mortality after knee revision . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 210
3.3.10 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 210
3.4 Outcomes after ankle replacement 213
3.4.1 Overview of primary ankle replacement surgery . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 214
3.4.2 Revisions after primary ankle surgery . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 221
3.4.3 Mortality after primary ankle replacement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 226
3.4.4 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 228
3.5 Outcomes after elbow replacement 229
3.5.1 Overview of primary elbow replacement surgery . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 230
3.5.2 Revisions after primary elbow replacement surgery . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 242
3.5.3 Mortality after primary elbow replacement surgery . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 252
3.5.4 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 256
3.6 Outcomes after shoulder replacement 257
3.6.1 Overview of primary shoulder replacement surgery . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 258
3.6.2 Revisions after primary shoulder replacement surgery . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 276
3.6.3 Patient Reported Outcome Measures (PROMs) Oxford Shoulder Scores (OSS) associated
with primary shoulder replacement surgery . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 288
3.6.4 Mortality after primary shoulder replacement surgery . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 298
3.6.5 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 303
3.7 In-depth studies 305
3.7.1 The effect of surgical approach in total hip replacement on outcomes: an analysis of
723,904 elective operations from the National Joint Registry for England, Wales, Northern
Ireland and the Isle of Man . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 306
3.7.2 What are the inpatient and day case costs following primary total hip replacement of
patients treated for prosthetic joint infection: a matched cohort study using linked data from
the National Joint Registry and Hospital Episode Statistics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 311
3.7.3 Effect of Bearing Surface on Survival of Cementless and Hybrid Total Hip Arthroplasty:
Study of Data in the National Joint Registry for England, Wales, Northern Ireland and
the Isle of Man . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 314
3.7.4 Provision of revision knee surgery and calculation of the effect of a network service
reconfiguration: An analysis from the National Joint Registry for England, Wales,
Northern Ireland and the Isle of Man . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 319
3.7.5 The association between surgical volume and failure of primary total hip replacement in
England and Wales: Findings from a prospective national joint replacement register . . . . . . . . . . . 324
Published papers 2020-2021 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 330
4. Implant and unit-level activity and outcomes 333
4.1 Implant performance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 334
4.2 Clinical activity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 336
4.3 Outlier units for 90-day mortality and revision rates for the period 2011 to 2021 . . . . . . . . . . . . . . . 337
4.4 Better than expected performance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 340
The effects of the COVID-19 pandemic on joint replacement surgery volumes and
waiting lists 341
The COVID-19 induced joint replacement deficit in England, Wales and Northern Ireland . . . . . . . . . . . . . . . . . . 342
The effects of the COVID-19 pandemic on joint replacement surgery: The patient perspective . . . . . . . . . . . . . . 358
Glossary 364
Infographic 374
Tables
3.1 Summary of data sources, linkage and methodology
Table 3.D1 Percentage data quality audit compliance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38
3.2 Outcomes after hip replacement
Table 3.H1 Number and percentage of primary hip replacements by fixation and bearing . . . . . . . . . . . . . . . . . . 47
Table 3.H2 Percentage of primary hip replacements by fixation, bearing and calendar year . . . . . . . . . . . . . . . . . 51
Table 3.H3 Age at primary hip replacement by fixation and bearing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60
Table 3.H4 Primary hip replacement patient demographics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61
Table 3.H5 KM estimates of cumulative revision (95% CI) by fixation and bearing, in primary hip replacements . . 65
Table 3.H6 KM estimates of cumulative revision (95% CI) of primary hip replacements by gender, age
group, fixation and bearing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 72
Table 3.H7 KM estimates of cumulative revision (95% CI) of primary hip replacement by fixation, and
stem / cup brand . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92
Table 3.H8 KM estimates of cumulative revision (95% CI) of primary hip replacement by fixation,
stem / cup brand, and bearing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 96
Table 3.H9 PTIR estimates of indications for hip revision (95% CI) by fixation and bearing . . . . . . . . . . . . . . . . . 101
Table 3.H10 PTIR estimates of indications for hip revision (95% CI) by years following primary hip replacement . . 103
Table 3.H11 KM estimates of cumulative mortality (95% CI) by age and gender, in primary hip replacement . . . 109
Table 3.H12 Number and percentage fractured NOF in the NJR by year . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 111
Table 3.H13 Fractured NOF vs. OA only by gender, age and fixation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 112
Table 3.H14 Number and percentage of hip revisions by procedure type and year . . . . . . . . . . . . . . . . . . . . . . 117
Table 3.H15 (a) Number and percentage of hip revision by indication and procedure type . . . . . . . . . . . . . . . . 118
Table 3.H15 (b) Number and percentage of hip revision by indication and procedure type in last five years . . . . 118
Table 3.H16 (a) KM estimates of cumulative re-revision (95% CI) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 127
Table 3.H16 (b) KM estimates of cumulative re-revision (95% CI) by years since first failure . . . . . . . . . . . . . . . . 127
Table 3.H16 (c) KM estimates of cumulative re-revision (95% CI) by fixation and bearing used in primary
hip replacement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 128
Table 3.H17 (a) Number of revisions by indication for all revisions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 129
Table 3.H17 (b) Number of revisions by indication for first linked revision and second linked re-revision . . . . . . 129
Table 3.H18 (a) Number of revisions by year . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 130
Table 3.H18 (b) Number of revisions by year, stage, and whether or not primary is in the NJR . . . . . . . . . . . . . 131
3.3 Outcomes after knee replacement
Table 3.K1 Number and percentage of primary knee replacements by fixation, constraint and bearing . . . . . . . 139
Table 3.K2 Percentage of primary knee replacements by fixation, constraint, bearing and calendar year . . . . . . 143
Table 3.K3 Age at primary knee replacement by fixation, constraint and bearing type . . . . . . . . . . . . . . . . . . . . 146
Table 3.K4 Primary knee replacement patient demographics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 147
Table 3.K5 KM estimates of cumulative revision (95% CI) by fixation, constraint and bearing, in primary
knee replacements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 151
Table 3.K6 KM estimates of cumulative revision (95% CI) by gender, age, fixation, constraint and
bearing, in primary knee replacements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 159
Table 3.K7 (a) KM estimates of cumulative revision (95% CI) by total knee replacement brands . . . . . . . . . . . . 167
Table 3.K7 (b) KM estimates of cumulative revision (95% CI) in total knee replacement brands by
whether a patella component was recorded . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 169
Table 3.K8 KM estimates of cumulative revision (95% CI) by unicompartmental knee replacement brands . . . . 174
Table 3.K9 (a) KM estimates of cumulative revision (95% CI) by fixation, constraint and brand . . . . . . . . . . . . . 175
Table 3.K9 (b) KM estimates of cumulative revision (95% CI) by fixation, constraint, brand and whether a
patella component was recorded . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 179
Table 3.K10 PTIR estimates of indications for revision (95% CI) by fixation, constraint, bearing type and
whether a patella component was recorded . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 187
Table 3.K11 PTIR estimates of indications for revision (95% CI) by years following primary knee replacement . . 190
Table 3.K12 (a) KM estimates of cumulative mortality (95% CI) by age and gender, in primary TKR . . . . . . . . . 191
Table 3.K12 (b) KM estimates of cumulative mortality (95% CI) by age and gender, in primary
unicompartmental replacements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 192
Table 3.K13 Number and percentage of revisions by procedure type and year . . . . . . . . . . . . . . . . . . . . . . . . . 194
Table 3.K14 (a) Number and percentage of knee revision by indication and procedure type . . . . . . . . . . . . . . . 195
Table 3.K14 (b) Number and percentage of knee revision by indication and procedure type in the last five years . 195
Table 3.K15 (a) KM estimates of cumulative re-revision (95% CI) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 206
Table 3.K15 (b) KM estimates of cumulative re-revision (95% CI) by years since first revision . . . . . . . . . . . . . . 206
Table 3.K15 (c) KM estimates of cumulative re-revision (95% CI) by fixation and constraint and whether a
patella component was recorded . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 207
Table 3.K16 (a) Number of revisions by indication for all revisions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 208
Table 3.K16 (b) Number of revisions by indication for first linked revision and second linked re-revision . . . . . . 208
Table 3.K17 (a) Number of revisions by year . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 209
Table 3.K17 (b) Number of revisions by year, stage, and whether or not primary is in the NJR . . . . . . . . . . . . . 210
3.4 Outcomes after ankle replacement
Table 3.A1 Descriptive statistics of ankle procedures performed by consultant and unit by year of surgery . . . . 219
Table 3.A2 Number and percentage of primary ankle replacements by ankle brand . . . . . . . . . . . . . . . . . . . . . 220
Table 3.A3 KM estimates of cumulative revision (95% CI) of primary ankle replacement, by gender and age . . . 221
Table 3.A4 KM estimates of cumulative revision (95% CI) of primary ankle replacement, by brand . . . . . . . . . . 223
Table 3.A5 Indications for the first revisions following primary ankle replacement . . . . . . . . . . . . . . . . . . . . . . . . 225
Table 3.A6 KM estimates of cumulative mortality (95% CI) after primary ankle replacement, by gender and age . . 227
3.5 Outcomes after elbow replacement
Table 3.E1 Number of primary elbow replacements by year and percentage of each type of procedure . . . . . . 233
Table 3.E2 Types of primary elbow procedures used in acute trauma and elective cases by year and
type of primary operation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 234
Table 3.E3 Indications for main confirmed types of primary elbow replacements, by year and type of
primary operation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 237
Table 3.E4 Number of units and consultant surgeons providing primary elbow replacements during each
year from the last three years, by region . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 238
(a) All primary elbow replacements (including the confirmed and unconfirmed total, radial head,
lateral resurfacing and distal humeral hemiarthroplasty replacements) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 238
(b) All confirmed primary total elbow replacements (with or without radial head replacement) . . . . . . . . . . . 239
Table 3.E5 Brands used in elbow replacement by confirmed procedure type . . . . . . . . . . . . . . . . . . . . . . . . . . 240
Table 3.E6 KM estimates of cumulative revision (95% CI) by primary elbow procedures for acute trauma
and elective cases . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 243
Table 3.E7 KM estimates of cumulative revision (95% CI) for all primary elbow procedures by implant brand . . 250
Table 3.E8 Indications for first data linked revision after any primary elbow replacement . . . . . . . . . . . . . . . . . . 251
Table 3.E9 KM estimates of cumulative mortality (95% CI) by time from primary elbow replacement, for
acute trauma and elective cases . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 253
3.6 Outcomes after shoulder replacement
Table 3.S1 Number and percentage of primary shoulder replacements (elective or acute trauma), by year
and type of shoulder replacement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 260
Table 3.S2 Demographic characteristics of patients undergoing primary shoulder replacements, by acute
or elective indications and type of shoulder replacement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 265
Table 3.S3 Numbers of units and consultant surgeons providing primary shoulder replacements and
median and interquartile range of procedures performed by unit and consultant, by year, last five years
and overall . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 266
Table 3.S4 Number and percentage of primary shoulder replacements by indication and type of shoulder
replacement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 267
Table 3.S5 (a) Number of resurfacing proximal humeral hemiarthroplasty replacements between 2012
and 2020 and within the last year by brand construct . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 268
Table 3.S5 (b) Number of stemless proximal humeral hemiarthroplasty replacements between 2012 and
2020 and within the last year by brand construct . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 268
Table 3.S5 (c) Number of stemmed proximal humeral hemiarthroplasty replacements between 2012 and
2020 and within the last year by brand construct . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 269
Table 3.S5 (d) Number of resurfacing total shoulder replacement replacements between 2012 and 2020
and within the last year by brand construct . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 270
Table 3.S5 (e) Number of stemless conventional total shoulder replacement replacements between 2012
and 2020 and within the last year by brand construct . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 271
Table 3.S5 (f) Number of stemmed conventional total shoulder replacements between 2012 and 2020
Table 3.S5 (g) Number of stemless reverse polarity total shoulder replacements between 2012 and 2020
Table 3.S5 (h) Number of stemmed reverse polarity total shoulder replacement replacements between
2012 and 2020 and within the last year by brand construct . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 274
Table 3.S6 KM estimates of cumulative revision (95% CI) for primary shoulder replacement for all cases,
acute trauma and elective cases . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 277
Table 3.S7 KM estimates of cumulative revision (95% CI) for primary shoulder replacement for elective
cases by gender and age group . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 278
cases by shoulder type . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 280
cases by brand construct in constructs with greater than 250 implantations . . . . . . . . . . . . . . . . . . . . . . . . . . . . 282
Table 3.S10 PTIR estimates of indications for shoulder revision (95% CI) for acute trauma by type of
shoulder replacement between 2012 and 2020 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 284
Table 3.S11 PTIR estimates of indications for shoulder revision (95% CI) for acute trauma by type of
shoulder replacement using reports from MDSv7 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 285
Table 3.S12 PTIR estimates of indications for shoulder revision (95% CI) for elective procedures by type
of shoulder replacement between 2012 and 2020 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 286
Table 3.S13 PTIR estimates of indications for shoulder revision (95% CI) for elective procedures by type
of shoulder replacement using reports from MDSv7 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 287
Table 3.S14 Number and percentage of patients who completed an Oxford Shoulder Score by acute
trauma and elective indications, by the collection window of interest at different time points . . . . . . . . . . . . . . . . 289
Table 3.S15 Number and percentage of patients who completed cross-sectional Oxford Shoulder
Score by overall, acute trauma, elective and by year of primary operation, within the collection window of
interest, with valid measurements at the time points of interest . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 291
Table 3.S16 Number and percentage of patients who completed longitudinal Oxford Shoulder Score by
overall, acute trauma, elective and by year of primary operation, within the collection window of interest,
with valid measurements at the time points of interest . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 292
Table 3.S17 Descriptive statistics of the pre-operative, 6-month and the change in OSS by overall, acute
trauma, elective and by year of primary operation, within the collection window of interest, with valid
measurements pre-operatively and 6 months post-operatively . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 295
trauma, elective and by shoulder type, within the collection window of interest, with valid measurements
pre-operatively and 6 months post-operatively . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 296
Table 3.S19 KM estimates of cumulative mortality (95% CI) by acute trauma and elective indications for
patients undergoing primary shoulder replacement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 299
Table 3.S20 KM estimates of cumulative mortality (95% CI) for primary shoulder replacement for elective
cases by gender and age group . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 302
3.7 In-depth studies
Table 3.1 Regression models to compare approach groups for revision risk . . . . . . . . . . . . . . . . . . . . . . . . . . . 308
(i) Cox proportional hazards regression models, with stratification by age/sex/risk groups . . . . . . . . . . . . . . 308
(ii) FPM models, with adjustment for time-varying effects of age, sex, risk group . . . . . . . . . . . . . . . . . . . . . 308
Table 3.2 Cox ‘proportional hazards’ regression model to compare 90-day mortality between the seven
approach sub-groups . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 309
Table 3.3 Characteristics of matched patients revised and not revised for PJI following primary THR . . . . . . . . 312
Table 3.4 Average total and annual inpatient and day case hospital admission costs over the five years
following THR, by revised PJI and comparator patients . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 313
Table 3.5 Cox regression hazard ratios of risk of any revision by bearing combination . . . . . . . . . . . . . . . . . . . . 317
Table 3.6 Cox regression hazard ratios of risk of any revision by bearing combination and head size . . . . . . . . . 318
Table 3.7 Revision surgeon volume, average annual number of revisions (2016-2018) . . . . . . . . . . . . . . . . . . . . 321
Table 3.8 The effect of MRC-RU reconfiguration on centre volume . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 322
4. Implant and unit-level activity and outcomes
Table 4.1 Level 1 outlier stems reported to MHRA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 334
Table 4.2 Level 1 outlier acetabular components reported to MHRA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 335
Table 4.3 Level 1 outlier stem/cup combinations reported to MHRA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 335
Table 4.4 Level 1 outlier implants reported to MHRA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 336
Table 4.5 Outliers for hip mortality rates since 2016 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 338
Table 4.6 Outliers for knee mortality rates since 2016 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 338
Table 4.7 Outliers for hip revision rates, all linked primaries from 2011 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 338
Table 4.8 Outliers for hip revision rates, all linked primaries from 2016 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 338
Table 4.9 Outliers for knee revision rates, all linked primaries from 2011 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 339
Table 4.10 Outliers for knee revision rates, all linked primaries from 2016 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 339
Table 4.11 Better than expected hip revision rates, all linked primaries from 2011 . . . . . . . . . . . . . . . . . . . . . . . 340
Table 4.12 Better than expected hip revision rates, all linked primaries from 2016 . . . . . . . . . . . . . . . . . . . . . . . 340
Table 4.13 Better than expected knee revision rates, all linked primaries from 2011 . . . . . . . . . . . . . . . . . . . . . 340
Table 4.14 Better than expected knee revision rates, all linked primaries from 2016 . . . . . . . . . . . . . . . . . . . . . 340
The COVID-19 induced joint replacement deficit in England, Wales

and Northern Ireland
Table 1 Descriptive statistics of provision and change of joint replacement by joint and nation . . . . . . . . . . . . . . 346
Table 2 Predicted years-to-recovery of 2020 deficit following expansion of joint replacement provision
compared to 2019 by joint type and nation. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 353
Figures
3.1 Summary of data sources, linkage and methodology
Figure 3.D1 Compliance rates from 2003 to 2014 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36
Figure 3.D2 Schematic presentation of NJR data compliance audit . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
Figure 3.D3 Initial numbers of procedures for analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39
3.2 Outcomes after hip replacement
Figure 3.H1 (a) Hip cohort flow diagram . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46
Figure 3.H1 (b) Frequency of primary hip replacements within elective cases stratified by procedure type . . . . . 48
Figure 3.H1 (c) Frequency of primary hip replacements within acute trauma cases stratified by procedure type . . . 49
Figure 3.H2 (a) Fixation and type by year of primary hip replacement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53
Figure 3.H2 (b) Unipolar THR fixation and main bearing type by year of primary hip replacement . . . . . . . . . . . . 54
Figure 3.H3 (a) Cemented primary hip replacement bearing surface by year . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55
Figure 3.H3 (b) Uncemented primary hip replacement bearing surface by year . . . . . . . . . . . . . . . . . . . . . . . . . . 56
Figure 3.H3 (c) Hybrid primary hip replacement bearing surface by year . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57
Figure 3.H3 (d) Reverse hybrid primary hip replacement bearing surface by year . . . . . . . . . . . . . . . . . . . . . . . . 58
Figure 3.H3 (e) Trends in fixation, bearing and head size in primary unipolar total hip replacement by year . . . . . 59
Figure 3.H4 (a) KM estimates of cumulative revision by year, in primary hip replacements . . . . . . . . . . . . . . . . . . 62
Figure 3.H4 (b) KM estimates of cumulative revision by year, in primary hip replacements plotted by year
of primary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63
Figure 3.H5 KM estimates of cumulative revision in cemented primary hip replacements by bearing . . . . . . . . . . 66
Figure 3.H6 KM estimates of cumulative revision in uncemented primary hip replacements by bearing . . . . . . . . 67
Figure 3.H7 KM estimates of cumulative revision in hybrid primary hip replacements by bearing . . . . . . . . . . . . . 68
Figure 3.H8 KM estimates of cumulative revision in reverse hybrid primary hip replacements by bearing . . . . . . . 69
Figure 3.H9 (a) KM estimates of cumulative revision in all primary hip replacements by gender and age . . . . . . . 70
Figure 3.H9 (b) KM estimates of cumulative revision in all primary hip replacements by gender and age,
excluding MoM and resurfacing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71
Figure 3.H10 (a) KM estimates of cumulative revision of primary cemented MoP hip replacement by
head size (mm) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80
Figure 3.H10 (b) KM estimates of cumulative revision of primary cemented CoP hip replacement by
head size (mm) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81
Figure 3.H10 (c) KM estimates of cumulative revision of primary uncemented MoP hip replacement by
head size (mm) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 82
Figure 3.H10 (d) KM estimates of cumulative revision of primary uncemented MoM hip replacement by
head size (mm) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83
Figure 3.H10 (e) KM estimates of cumulative revision of primary uncemented CoP hip replacement by
head size (mm) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84
Figure 3.H10 (f) KM estimates of cumulative revision of primary uncemented CoC hip replacement by
head size (mm) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85
Figure 3.H10 (g) KM estimates of cumulative revision of primary hybrid MoP hip replacement by head size (mm) . 86
Figure 3.H10 (h) KM estimates of cumulative revision of primary hybrid CoP hip replacement by head size (mm) . . 87
Figure 3.H10 (i) KM estimates of cumulative revision of primary hybrid CoC hip replacement by head size (mm) . . 88
Figure 3.H10 (j) KM estimates of cumulative revision of primary reverse hybrid MoP hip replacement by
head size (mm) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89
Figure 3.H10 (k) KM estimates of cumulative revision of primary reverse hybrid CoP hip replacement by
head size (mm) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 90
Figure 3.H10 (l) KM estimates of cumulative revision of primary resurfacing MoM hip replacement by
head size (mm) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91
Figure 3.H11 (a) PTIR estimates of aseptic loosening by fixation and bearing . . . . . . . . . . . . . . . . . . . . . . . . . . 105
Figure 3.H11 (b) PTIR estimates of pain by fixation and bearing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 105
Figure 3.H11 (c) PTIR estimates of dislocation / subluxation by fixation and bearing . . . . . . . . . . . . . . . . . . . . . 106
Figure 3.H11 (d) PTIR estimates of infection by fixation and bearing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 106
Figure 3.H11 (e) PTIR estimates of lysis by fixation and bearing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 107
Figure 3.H11 (f) PTIR estimates of adverse soft tissue reaction by fixation and bearing . . . . . . . . . . . . . . . . . . . 107
Figure 3.H11 (g) PTIR estimates of adverse soft tissue reaction by fixation and bearing, since 2008 . . . . . . . . . 108
Figure 3.H12 (a) KM estimates of cumulative revision for fractured NOF and OA only cases for primary
hip replacements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 113
Figure 3.H12 (b) KM estimates of cumulative revision by bearing type for fractured NOF cases in primary
hip replacements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 114
Figure 3.H13 KM estimates of cumulative mortality for fractured NOF and OA only in primary hip replacements . . 115
Figure 3.H14 (a) KM estimates of cumulative re-revision in linked primary hip replacements . . . . . . . . . . . . . . . 119
Figure 3.H14 (b) KM estimates of cumulative re-revision by primary fixation in linked primary hip replacements . . 120
Figure 3.H14 (c) KM estimates of cumulative re-revision by years to first revision, in linked primary hip
replacements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 121
Figure 3.H15 (a) KM estimates of cumulative re-revision in cemented primary hip replacement by years
to first revision, in linked primary hip replacements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 122
Figure 3.H15 (b) KM estimates of cumulative re-revision in uncemented primary hip replacement by
years to first revision, in linked primary hip replacements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 123
Figure 3.H15 (c) KM estimates of cumulative re-revision in hybrid primary hip replacement by years to
first revision, in linked primary hip replacements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 124
Figure 3.H15 (d) KM estimates of cumulative re-revision in reverse hybrid primary hip replacement by
years to first revision, in linked primary hip replacements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 125
Figure 3.H15 (e) KM estimates of cumulative re-revision in resurfacing primary hip replacement by years
to first revision, in linked primary hip replacements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 126
3.3 Outcomes after knee replacement
Figure 3.K1 (a) Knee cohort flow diagram . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 137
Figure 3.K1 (b) Frequency of primary TKR within elective cases stratified by procedure type . . . . . . . . . . . . . . 140
Figure 3.K1 (c) Frequency of primary UKR within elective cases stratified by procedure type . . . . . . . . . . . . . . 141
Figure 3.K1 (d) Frequency of primary patellofemoral knee replacements within elective cases stratified by
procedure type . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 142
Figure 3.K2 Fixation by year of procedure in primary knee replacement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 145
Figure 3.K3 (a) KM estimates of cumulative revision by year, in primary knee replacements . . . . . . . . . . . . . . . 148
Figure 3.K3 (b) KM estimates of cumulative revision by year, in primary knee replacements plotted by
year of primary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 150
Figure 3.K4 (a) KM estimates of cumulative revision in primary total cemented knee replacements by
constraint and bearing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 153
Figure 3.K4 (b) KM estimates of cumulative revision in primary total uncemented knee replacements by
Figure 3.K4 (c) KM estimates of cumulative revision in primary total hybrid knee replacements by
Figure 3.K4 (d) KM estimates of cumulative revision in primary unicondylar or patellofemoral knee
replacements by fixation, constraint and bearing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 156
Figure 3.K5 (a) KM estimates of cumulative revision in primary total knee replacements by gender and age . . . 157
Figure 3.K5 (b) KM estimates of cumulative revision in primary unicondylar knee replacements by gender
and age . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 158
Figure 3.K6 (a) KM estimates of cumulative re-revision, in linked primary knee replacements . . . . . . . . . . . . . . 197
Figure 3.K6 (b) KM estimates of cumulative re-revision by primary fixation, in linked primary knee replacements . 198
Figure 3.K6 (c) KM estimates of cumulative re-revision by years to first revision, in linked primary knee
replacements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 199
Figure 3.K7 (a) KM estimates of cumulative re-revision in primary cemented TKRs by years to first revision . . . 200
Figure 3.K7 (b) KM estimates of cumulative re-revision in primary uncemented TKRs by years to first revision . . . 201
Figure 3.K7 (c) KM estimates of cumulative re-revision in primary hybrid TKRs by years to first revision . . . . . . 202
Figure 3.K7 (d) KM estimates of cumulative re-revision in primary patellofemoral knee replacements by
years to first revision . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 203
Figure 3.K7 (e) KM estimates of cumulative re-revision in primary cemented unicondylar knee
replacements by years to first revision . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 204
Figure 3.K7 (f) KM estimates of cumulative re-revision in primary uncemented / hybrid unicondylar knee
replacements by years to first revision . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 205
3.4 Outcomes after ankle replacement
Figure 3.A1 Ankle cohort flow diagram . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 215
Figure 3.A2 Fixation by year of primary ankle replacement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 216
Figure 3.A3 Frequency of primary ankle replacements, bars stacked by volume per consultant per year . . . . . . 217
Figure 3.A4 Frequency of primary ankle replacements stratified by fixed and mobile bearings, bars
stacked by volume per consultant per year . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 218
Figure 3.A5 KM estimates of cumulative revision of primary ankle replacement . . . . . . . . . . . . . . . . . . . . . . . . . 222
Figure 3.A6 KM estimates of cumulative revision of primary ankle replacement by brand . . . . . . . . . . . . . . . . . 224
Figure 3.A7 KM estimates of cumulative mortality after primary ankle replacement . . . . . . . . . . . . . . . . . . . . . . 226
3.5 Outcomes after elbow replacement
Figure 3.E1 Elbow cohort flow diagram . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 231
Figure 3.E2 Frequency of primary elbow replacements within elective cases stratified by procedure type,
bars stacked by volume per consultant per year . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 235
Figure 3.E3 Frequency of primary elbow replacements within acute trauma cases stratified by procedure
type, bars stacked by volume per consultant per year . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 236
Figure 3.E4 KM estimates of cumulative revision of primary elbow replacement by acute trauma and
elective cases . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 245
Figure 3.E5 KM estimates of cumulative revision of primary total elbow replacement (with or without a
radial head replacement) by acute trauma and elective cases . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 246
Figure 3.E6 KM estimates of cumulative revision of primary radial head replacement by acute trauma and
elective cases . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 247
Figure 3.E7 KM estimates of cumulative revision of total elbow replacements and distal humeral
hemiarthroplasty within the acute trauma cases . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 248
Figure 3.E8 KM estimates of cumulative revision of total elbow replacements by implant brand within the
elective cases . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 249
Figure 3.E9 KM estimates of cumulative mortality of primary elbow replacement by acute trauma and
elective cases . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 255
3.6 Outcomes after shoulder replacement
Figure 3.S1 Shoulder cohort flow diagram . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 259
Figure 3.S2 Frequency of primary shoulder replacements within elective patients stratified by procedure
type, bars stacked by volume per consultant per year . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 262
Figure 3.S3 Frequency of primary shoulder replacements within acute trauma patients stratified by
procedure type, bars stacked by volume per consultant per year . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 263
Figure 3.S4 Age (Box and whiskers) and frequency of primary shoulder replacements by gender and
type of shoulder replacement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 264
Figure 3.S5 KM estimates of cumulative revision for primary shoulder replacement by acute trauma and
elective cases . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 277
Figure 3.S6 KM estimates of cumulative revision for primary elective shoulder replacement by type of
shoulder replacement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 279
Figure 3.S7 KM estimates of cumulative revision for elective primary elective shoulder replacements for
patients with and without valid PROMs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 293
Figure 3.S8 Distribution and scatter of pre-operative OSS and the change in OSS (post-pre) score for
those receiving elective shoulder replacements for valid measurements within the collection window of interest . . . 294
Figure 3.S9 KM estimates of cumulative mortality by acute trauma and elective indications for patients
undergoing primary shoulder replacement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 298
Figure 3.S10 KM estimates of cumulative mortality for primary elective shoulder replacement by gender . . . . . 300
Figure 3.S11 KM estimates of cumulative mortality for primary elective shoulder replacement by age
group and gender . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 301
3.7 In-depth studies
Figure 3.1 Cumulative percentage revised (Kaplan-Meier) up to 12 years for the seven surgical approach
groups . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 307
Figure 3.2 Cumulative incidence of revision for any reason by bearing combination . . . . . . . . . . . . . . . . . . . . . . 315
Figure 3.3 Cumulative incidence of revision for any reason by bearing combination in patients aged less
than 55 years of age at the time of primary THR . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 316
Figure 3.4 Individual surgeon volume . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 320
Figure 3.5 Site versus surgeon volume . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 321
Figure 3.6 Empirical cumulative distribution and frequency distribution of (between) mean consultant
volume and (within) individual centred volume of hip arthroplasty in the previous 365 days . . . . . . . . . . . . . . . . . 325
Figure 3.7 Mean, IQR and 95th centile range of consultant and centre volume of hip arthroplasty in the
previous 365 days recorded in the NJR by individual consultant and individual unit, respectively . . . . . . . . . . . . 326
Figure 3.8 Between-consultant marginal association of hip surgical volume in the preceding 365 days
and hazard of revision arthroplasty unadjusted and adjusted for confounding factors in a multilevel model . . . . . 327
Figure 3.9 Within-consultant marginal association of hip surgical volume in the preceding 365 days and
hazard of revision arthroplasty unadjusted and adjusted for confounding factors in a multilevel model . . . . . . . . 328
The COVID-19 induced joint replacement deficit in England, Wales

and Northern Ireland
Figure 1 Annual number of primary hip, knee, shoulder, elbow and ankle replacements performed in
England, Wales and Northern Ireland . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 347
Figure 2 Weekly number of primary hip and knee replacements performed in England, Wales, and
Northern Ireland in 2019 and 2020 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 348
Figure 3 Weekly number of primary shoulder, elbow and ankle replacements performed in England,
Wales, and Northern Ireland in 2019 and 2020 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 349
Figure 4 Weekly number of primary shoulder, elbow and ankle replacements performed in
England,Wales, and Northern Ireland in 2020 by nation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 350
Figure 5 Predicted years-to-recovery of the 2020 deficit of joint replacement procedures following
expansion of joint replacement provision compared to 2019 in England, Wales and Northern Ireland . . . . . . . . . 351
Figure 6 Predicted years-to-recovery of the 2020 deficit of joint replacement procedures following
expansion of joint replacement provision compared to 2019 stratified by nation . . . . . . . . . . . . . . . . . . . . . . . . . 352
1. Chairman’s
Foreword
Chairman’s Foreword
Laurel Powers-Freeling
Chairman, National Joint Registry Steering Committee (NJRSC)
The NJRSC oversees the strategic and operational

work programme of the registry and I am delighted to
have performed the role of Chairman of the Committee
over the past ten years, which means, following good
governance standards, it is time for me to make way
for a new person in the NJR chair.
In each of the past ten years, I have had exciting

news to share regarding the evolution of the NJR.
While this year has been very challenging in the wake
of the pandemic - the NJR nonetheless delivered a
number of important developments. This NJR Annual
Report provides the opportunity to reflect back on our
work over the last year and look to the year ahead.
Highlights are summarised here in this 18th edition of
our Annual Report. publication of the Cumberlege report, which identified
the need for more comprehensive implant device
Our work and developments registries and cited the NJR as a ‘global exemplar’
of such a registry. This provided the opportunity for
Managing the impact of the COVID-19 crisis: us to propose the MSK registry be considered as a
In 2020/21 the NJR undertook a radical review of useful pilot for plans to develop appropriate options
our proposed annual work plan and budget to reflect for implementation of larger integrated data sets for
the impact of the COVID-19 crisis. We considered implantable devices. We will pursue this objective in
how resources could be conserved until we could re- the coming year and ensure that we continue to align
engage in collecting, processing and analysing data with national plans to deliver appropriate options for
for our work and reinstating income collection via trust implementation of a centralised registries database.
subscription payments. As a result, our development
plans and expenditure programme last year were Automating our Data Quality Audit: Data quality
significantly reduced. We will continue to monitor our has continued to be a key priority for the NJR and
activity and finance in 2021/22, to ensure the impact our Data Quality Audit programme has been a unique
of reduced elective surgery and its effect on trust initiative with considerable success in assessing the
subscription income continues to be managed. completeness and quality of the data submitted to
the registry. However, the process of comparing
NJR/BOA plans for implementation of a local hospital records to those submitted to the NJR
Musculoskeletal [MSK] Registry: This year we have has been labour intensive for both hospital and NJR
continued to pursue the proposal to develop a national staff, so we began a national roll out of an enhanced,
MSK registry, bringing the seven registries forming automated process. This has greatly reduced the
the BOA Trauma and Orthopaedic Registries Unifying burden involved in undertaking this work and enables
Structure (TORUS) together with the NJR, under a units to check their data quality on a more frequent
single governance body. This proposal has gained basis. Full roll out of this enhanced process in all
support in principle from NHS leadership following joint types was completed in 2020/2021. In addition,
data quality exercises involving dual mobility hip feedback systems, interactive reporting, availability of
replacements, reverse shoulder replacements and an implant data library, a new semantic layer to aid
multi-compartmental knee replacements are now researcher secure access to NJR data, data entry and
being developed in consultation with the relevant data quality tools, and the Data Access Portal.
specialist societies.
Unique/innovative solutions to support
PROMs and ePREMs: An important element of patient safety: Following the NHS Healthcare
work continues to be the collection, analysis and Safety Investigation Branch requirement to reduce
reporting, of patient reported metrics and this year the number of ‘never events’ associated with joint
we have focused on a number of key areas. We have replacement surgery, the NJR has been working
been examining the quality and representativeness of to deliver validation rules that apply in data entry to
national PROMs for patients who have hip and knee an external environment, for use in support of intra-
replacement surgery, which will ensure that clinicians, operative checks. The data entry system has been
hospital managers and regulators can have confidence updated to enable it to detect potential ‘never events’
in using patient reported metrics to assure quality. and warn the data entry user, also alerting the NJR
We have also been routinely including reporting of team so that this can be investigated. We have also
PROMs metrics in implant reports made available to developed an Application Programming Interface, to
manufacturers via the NJR Supplier Feedback service allow hospital theatre systems to interface with NJR’s
and developing a system that will make a library of checking rules and enable immediate identification of
these implant reports available to clinical teams via a implant incompatibilities as ‘never events’ in real time,
new digital platform that will be launched in 2021/22. so these are identified before the implant is put in the
In conjunction with both BESS and GIRFT, we have patient. A smartphone version of this application is
been working to improve patient engagement with also being developed so clinical teams can undertake
pre-operative PROMs submissions and in the coming validation checks even if their hospital does not have a
year we will commence an electronic Patient Reported compatible front-end system. This work also supports
Experience Measures (ePREMs) pilot, halted this year the importance of the emphasis on patient safety as
due to the COVID crisis, where we hope that patients highlighted in the Cumberlege report.
will share their experiences of joint surgery to help
improve healthcare for patients. Research and the NJR Data Access Portal (DAP):
Research has been a huge part of the NJR’s success
Modernising our IT Platform – Launch of ‘NJR and the output of peer-reviewed papers by the
CONNECT’: Last year we commissioned the University of Bristol and by others using NJR data, has
development of a cloud-ready, platform-based been truly extraordinary and ensures that NJR data
application framework for provision of future NJR can be best used to inform and improve practice. It
services. The rationale for this included a focus on has led to a large number of important and impactful
the need to develop an environment with the ability publications, delivering valuable evidence about how
to move to a cloud-based infrastructure and have joint replacement surgery works and with the key aim
the capacity to extend to any additional registry of being used to improve patient safety and outcomes.
alignment. This year the first phase of development This year the NJR DAP has been developed to
launched and transferred the NJR Clinician Feedback streamline research applications by providing a secure
services into the new environment, along with a more working environment, including analysis tools for
interactive reporting service, including the Consultant researchers and users of NJR data, whilst enabling the
and Surgeon Level Report, Annual Clinical Report, NJR to manage and control our data more effectively.
Clinical Outcomes Publication Preview, Clinician Profile Providing access to the data without the need for
Edit, interactive outcomes and clinical practice reports, datasets to be sent to third parties will significantly
and a contacts database. Further development reduce the governance burden that research
will continue during 2021/22 and include the NJR teams face.
component database and supplier and management
The NJR Patient Decision Support Tool: A major in data upload. In addition, the University of Oxford
initiative has been the launch of the NJR Patient have proposed a new component classification for
Decision Support Tool, a web-enabled personalised shoulder arthroplasty devices, which is planned to be
decision-making tool for patients considering hip or published on open access and will be free to license
knee replacement. This tool, whose development and will further form the basis of a new shoulder
was in collaboration with the University of Sheffield component database, to be developed by the NJR in
and supported by the charity Versus Arthritis, will help the coming year.
patients considering joint replacement make evidence-
based choices about their treatment and share The people who make the NJR
decision-making with their clinicians when considering a success
the benefits and risks of undergoing joint replacement.
We are continuing a collaboration with the University This year has seen a number of changes to the
of Sheffield to enhance the tool to allow the most up- NJRSC membership. I am delighted to welcome
to-date NJR data to be used to calculate the projected Derek Pegg as a co-opted member of the NJRSC, in
risks and benefits of joint replacement surgery. This his new role as Chairman of the NJR Regional Clinical
NJR initiative will continue to benefit healthcare Coordinators (RCC) and Data Quality Committees,
economies through improved clinical outcomes and succeeding Matthew Porteous. Derek has supported
better resource utilisation. the NJR for many years as Vice Chairman of the RCC
Committee and through membership of a number of
Redevelopment of the NJR Website: Work has NJR sub-committees. I thank him for his continued
been ongoing to design and build the architecture support and wish him well in his new roles. I am also
for our new website. As our public-facing information very pleased to confirm the re-appointment for a
portal, the aim of the upgrade is to develop increased further term of office, of both Peter Howard, NJRSC
functionality to make the website more engaging and surgeon member and Robin Brittain, NJRSC patient
enable us to develop new visual material to inform our representative member and to thank them for their
stakeholders more imaginatively about the work we continued hard work. My appreciation also goes to
do and to clearly demonstrate how the NJR benefits Bob Handley for his contribution as BOA President
the orthopaedic sector. The website is scheduled to to the NJRSC this year, which has been important in
launch during 2021/22. continuing our valued relationship with the orthopaedic
profession. We look forward to welcoming his
NJR Component Database and International
successor John Skinner, who takes up post from
Benefits: Following work with the German
September 2021.
Arthroplasty Registry (EPRD) to develop a common
classification system for defining the attributes of hip As ever, my grateful thanks go to the NJR Regional
and knee arthroplasty components, the classification Clinical Coordinators who underpin and champion
has been adopted by both our registries, each of us the work and success of the NJR at a local level. Also
managing our own local databases populated by to our contract partners Northgate Public Services
industry implant suppliers. Use of this classification (UK) Ltd (who will be known as NEC Software
data for NJR reporting will commence in 2021/22. In Solutions UK Ltd from July 2021) and the University
addition, we have agreed to license the component of Bristol, for their excellent work throughout the year
classification system to the International Society in supporting the NJR to deliver its work agenda
of Arthroplasty Registers for their International and objectives. I would like to end by thanking all
Prosthesis Library platform, meaning the hip and knee members of the NJRSC and sub-committees for their
components can be classified in the same way in valuable contribution. In particular, my thanks to Tim
registries across much of the world. This will provide Wilton, NJR Vice Chairman and Medical Director,
the valuable international benefits of improving the for his clinical expertise and leadership and for his
comparability of data for identifying poor outcomes interim chairmanship of the RCC and Data Quality
and of decreasing the burden on industry colleagues committees, pending the appointment of Derek Pegg.
My sincere thanks also to the Chairmen of each of

our sub-committees - Peter Howard, Mark Wilkinson,
Mike Reed and Derek Pegg - for their hard work, vision
and effort. Without their dedication, the NJR would not
be the world-leading arthroplasty register and global
exemplar of an implantable device registry that it is. I
would encourage you to read the reports from each
committee Chairman at reports.njrcentre.org.uk
where they provide strategic oversight into key
work areas.
Finally, my thanks as ever to the NJR Management

Team who in the past year have had to cope with
an environment of uncertainty in how and where
to work, with shifting priorities and compressed
budgets. They have continued to support the NJR
cheerfully and tirelessly, against a challenging
background. I particularly want to thank Elaine
Young, my partner in all things NJR for the past
decade, whose dedication to what the NJR delivers
has always been extraordinary.
I leave the NJR with mixed feelings: I am immensely

proud to have been associated with so many talented,
dedicated professionals and to have been a part
of a truly extraordinary organisation. But I am also
concerned about how the NJR we have all worked so
hard to build will fare in the headwinds of a challenging
NHS environment. Having said that, I know I leave our
organisation in the capable and protective hands of
my extraordinary Steering Committee colleagues…but
I will be watching!
Laurel Powers-Freeling
Chairman, National Joint Registry Steering Committee
2. Executive
Summary
Executive summary
Professor Mike Reed Mr Tim Wilton

Chairman, Editorial Board NJR Medical Director
It will come as no surprise to anyone that the single operation complexity changing, and alterations in
most pronounced factor in this year’s annual report readiness to perform both primary and revision
compared to previous years is the massive impact procedures during the pandemic.
of the COVID-19 pandemic on the volume of all joint
procedures. This has meant not only that the number The pandemic has of course shaped our representation
of cases performed in 2020 has been roughly halved at both national and international meetings, but many
across the whole spectrum of arthroplasty, but this have continued virtually. We supported both the BASK
fall in numbers occurred during the last nine months and BHS annual conferences this year with virtual
of the year. This means that actual loss of arthroplasty presentations from our Medical Director Tim Wilton
provision was closer to 70% during those nine months and Peter Howard, Chairman of our Implant and
and it is clear that volume has not fully recovered during Surgical Performance sub-committees. Each session
the first quarter of 2021. We can anticipate that the data was followed by a lively open question session with
for analysis will be distorted by this loss of throughput delegates who were interested in hearing how the NJR
and the accompanying altered case-mix for some is supporting the work of the orthopaedic sector.
years to come. Our preliminary analysis suggests that
Meanwhile the work of the NJR Editorial Board has
simply recovering the 2020 deficit will take a decade if
continued. The Board develops the strategy and style
joint surgery can only be increased by 5% compared
of the report and all members take responsibility for
to 2019; and will take five years if a 10% increase can
producing a report that is rigorously edited, taking
be achieved. Recovery will clearly take much longer at
almost a full year to write and review. The Board brings
those rates when the further deficit in volumes that has
together experts on data collection and reporting as
continued into 2021 is factored in.
well as generous input from a patient perspective,
Readers will therefore have to interpret much of the clinicians from specialist societies and members of the
data from all arthroplasty registries with great care NJR Management Team. Each year the Board aims to
over the coming years as there will have been multiple make progress in reporting on our rich data resource,
reasons why the outcome results may be different making data easily accessible to improve patient
from previous years including: patients waiting longer, outcomes. In addition to the section on COVID-19,
a key development for this report has been “volume hip replacements are performed by surgeons doing
plots” which show the number of specific procedures more than 97 such cases a year. Only a tiny number
performed each year, but also demonstrate whether of resurfacing hips are performed by surgeons doing
each procedure was performed by surgeons with more than 97 such operations, but the proportion
higher or lower activity. of resurfacing cases done by those surgeons is
nevertheless very high. This indicates that surgeons
We hope to launch the report at the British Orthopaedic who are performing such operations tend to be highly
Association Congress meeting in Aberdeen in specialised in the procedure.
September – and at the time of going to press we are
looking forward to this being a face-to-face meeting! Dual mobility hips, although performed far less
frequently, are on the rise with a steady climb – being
This year we will have a limited print run of the annual almost unheard of in 2013. It is perhaps not surprising,
report to be issued on a first-come first-served basis given the more limited indications for the procedure,
at the report launch. Increasingly there is considerable that even prior to the pandemic few surgeons were
additional information available online and we would performing more than 25 dual mobility hip procedures
encourage you to explore the NJR’s dedicated annual per year, but most such operations are nevertheless
report website at reports.njrcentre.org.uk. The performed by surgeons who do more than seven
website offers a helpful interactive platform for the per annum.
descriptive NJR data, with supporting appendices.
For the first time, hybrid fixation has become the most
Commentary on findings popular choice for hip replacement. It is interesting to

note that while for cemented and reverse hybrid fixation
metal-on-polyethylene (MoP) remains the predominant
This year NJR’s Annual Report is based on 2,895,368 bearing surface choice; in both uncemented and hybrid
records entered between 1 April 2003 and 31 hips the favoured choice is ceramic-on-polyethylene
December 2020, and the NJR maintains its position as (CoP). Over the years there has been little change in
the largest orthopaedic registry in the world. The report the choice of bearing surface for reverse hybrid hips,
presents joint replacement up to 17 years of follow-up, and a very gradual change from MoP towards CoP
with data on hips, knees, shoulders, elbows and ankle for cemented hips. In contrast the change to CoP
replacements. Due to the pandemic, approximately has been much more marked for hybrid hips and the
half as many records were added this year. In total the bearing choice in uncemented hips has seen far more
following numbers of linkable primary joint replacements pronounced variations over the years. The reasons for
are available for analysis: 1,251,164 hips, 1,357,077 this greater variation for uncemented hips is unclear and
knees, 7,084 ankles, 50,255 shoulders and 5,043 deserves further research and clarification.
elbow replacements. There are further linkable revisions
for each joint. The temporal changes in total hip replacement (THR)
revision (Figure 3.H4 (a)) indicate a deterioration
Hip replacement in revision rates until 2008 followed by marked
improvement at all time points. There may have been
There are new graphic representations of the two phases in this recovery; a sharp improvement
proportions of different hip operation types performed until 2010 followed by a more gradual improvement
by surgeons according to their annual throughput of ever since. It is tempting to suggest the steeper
those cases, and these give a fascinating insight into improvement from 2008 to 2010 is due mainly to
how things have changed over the years. Bearing in metal-on-metal (MoM) issues and the subsequent
mind the concerns over minimum numbers and low improvement is a secular trend which is also seen in
surgeon volumes, these graphs give useful information knee revision rates. The reason for that secular trend
about the general level of surgeon experience. High may be multi-factorial, but coincides with the start of
proportions of most types of hip replacement can feedback of data on their own results to surgeons by
be seen to be performed by surgeons doing quite the NJR. In Figure 3.H4 (b) the 13-year revision rates
high numbers per year. More than half of unipolar
are more clearly seen to be following the improving examined in more detail to see if there may have been
trends of earlier time points. a real drop in such fractures, or whether other factors
such as altered case-mix or altered threshold for certain
Revision rates for different revision indications are treatments were responsible. Revision rates for THR
shown in Figure 3.H11 and it is perhaps no surprise performed for hip fracture seem to track those for THR
that rates for aseptic loosening (and lysis) are highest for in osteoarthritis to a remarkable extent out to 15 years,
MoM bearings and resurfacing. Although very low rates albeit with increased rates of revision for the former in
for aseptic loosening are seen in cemented MoP and the first year.
CoP these do seem to rise significantly after ten years.
The very low rates seen in hybrid MoP and CoP up to Knee replacement
ten years do also appear to cross over such that hybrid
CoC have the lower rate for aseptic loosening after ten There are many areas this year in which there is either
years. There appear to be few differences in infection new information in the report or the previous information
rates according to bearing or fixation except for a higher is expanded to give much more detail. There has been
rate in MoM hips at most lengths of follow-up. This data controversy for years about the possibility that surgeons
is unadjusted however and two separate analyses from doing small numbers of certain operations may be
registry data have shown reduced infection rates in systematically giving rise to higher failure rates than
ceramic bearings. those performing higher numbers. This relationship
seems particularly clear in the case of unicondylar
Interesting detail is available this year on revision rates knee replacements and this year’s data show that the
by bearing surface stratified by age groups, and also median number of such cases performed over the
for femoral head size; and surgeons will want to delve past three years is 19 (per surgeon) or 49 per unit.
deeply into this information to check whether more or These figures show that, on average, surgeons are
less “tailoring” of their procedures may be desirable still not reaching the target numbers set by the British
according to the individual patient being treated. Association for Surgery of the Knee (BASK) which
will be of concern to many. Nevertheless, over recent
Revision rates (PTIRs) for different hip constructs are
years we can see that over two-thirds of unicondylar
presented in much more detail this year and should
procedures are performed by surgeons doing 25 cases
be of particular interest to both surgeons and patients
or more per year.
alike. It can be seen that some apparently similar
constructs have differing revision rates and surgeons Although the case numbers per surgeon have been
will wish to reassure themselves that what they may greatly distorted by the impact of COVID-19 in 2020,
believe about the construct they are using is indeed we can see that over recent years prior to 2020, virtually
borne out by this extensive analysis. Readers should no procedures were carried out by surgeons performing
also be aware that these revision rates for constructs fewer than seven total knee replacements (TKRs) per
are not adjusted for age and other case-mix variables, year and that about 75% of such cases each year were
but some of the constructs may be specifically by surgeons performing more than 49 cases per year.
indicated in younger patients or for some specific
indication, so it is necessary to look carefully at the age, Table 3.K1 shows that over the life of the registry
gender and other factors presented. roughly 25% of all TKRs are performed using a
posterior stabilised (PS) implant and these consistently
The median age and interquartile range are often quite show higher revision rates than operations performed
different between cemented, hybrid and uncemented with unconstrained implants. While it is sometimes
constructs, and this is most marked when considering argued that this could reflect PS usage by many
resurfacing constructs. surgeons when they encounter a particularly difficult or
complicated case, the evidence from the registry seems
In 2020 there was a marked reduction in THR for
to show that the choice is mostly based on surgeon
hip fracture and this might be the group where we
preference. This is therefore an area where surgeons
would have expected numbers to hold up despite
may wish to reflect on whether they are really making
the pandemic. The reasons for this will need to be
the safest and most appropriate choice.
The revision rates for PS and cruciate retaining (CR) wholly, or only in part, due to the different indications
knees of each implant brand are now shown in the report for first revision which predominate in the early and late
separately and are also stratified according to whether post-operative periods.
the patella is resurfaced or not. These data show that
the above finding of higher revision rates in PS knees Elbow replacement
may in fact be seen in the majority of, but not all, implant
brands. It is therefore worthwhile for surgeons to look at There are now over 5,000 elbow replacements
Figures 3.K7 to 3.K10 in detail to ascertain the precise available for analysis including total replacement (with
differences between the sub-types of implant available to or without radial head replacement), distal humeral
them in their units, and in the market in general. hemiarthroplasty, lateral resurfacing and radial head
replacement. Over 40% of these were performed for a
Although unicondylar knees are seen to have typically trauma indication.
involved the use of a mobile bearing over the last 18
years, this has been changing recently towards a fixed With the exception of 2020, the number of elbow
bearing. There is still controversy about the pros and replacements being registered has increased but the
cons of performing unicondylar knee procedures, and numbers of surgeons performing one to two per year
caution needs to be used in interpreting the revision has fallen, and those performing more than 13 has
data since there are clear case-mix differences between increased. Revision rates differ by indication, with
those suitable for a unicondylar and those who are not. primary total elbow replacement (with or without a radial
The statistics reported here have not been adjusted for head replacement) for acute trauma being less than 4%
such case-mix differences, which may be of particular at eight years, with elective indications being less than
relevance in certain implants which have been marketed 10% at eight years, although few cases have that length
as suitable for a particular patient grouping such as of follow-up.
younger, more active patients or a particular gender.
Shoulder replacement
As predicted in the last two years, Figure 3.K3 (b)
shows that the improvement in revision rates shown A rigorous review of the shoulder data has been
previously up to ten years, is now reflected at 13 and 15 performed. Consequently, new classifications and
years as well, starting as before with the 2008 cohort. component attributes are now used within the report
to define the primary groupings throughout the whole
While revision rates for uncemented unicondylar of the shoulder section. The report has now moved
procedures appear lower than those for cemented to whole construct validation, ensuring all relevant
at many time points, it is important to view this in the elements required to build a construct are present in
context of a gradual improvement in unicondylar results every procedure being reported on in our analysis. Over
in general, which has been occurring over 15 years. 50,000 primary shoulder replacements are available
Thus the uncemented unicondylar implants, which have for analysis. The proportion of reverse polarity total
mostly been inserted over the last ten years, would be shoulder replacement continues to increase (see Table
expected to have slightly better results than cemented 3.S1) albeit tending towards use in somewhat older
unicondylars with similar follow-up but performed on patients (Table 3.S4). The median, interquartile range,
average more than ten years ago. and number of procedures performed by units and
consultants has remained static for the last few years,
Re-revision rates continue to be seen to be much apart from 2020 due to the impact of COVID-19.
higher than those revised after primary operations
across all sub-groups of knee replacement. Figures Ankle replacement
for re-revision of around 16% at ten years are seen
this year, which is several times higher than those for This report focuses on primary procedures performed,
a standard primary TKR and therefore continue to be and also on revision and mortality, with over 7,000
a cause for concern. The time to first revision has a procedures being available for analysis. As noted
huge impact on the likelihood of subsequent revision previously, all ankle replacements recorded use
procedures, but it remains uncertain whether this is uncemented implants although cement was listed in the
component data in less than 5% and in the context of to address this gap in implant outcome reporting by
poor bone stock and low-demand patients. incorporating national PROMs analyses within our future
annual reports.
The proportion of fixed-bearing ankle replacements
continue to increase, and most procedures are done by Data will be analysed separately for hip and knee joints.
surgeons performing more than seven cases per year. Analyses will be repeated according to indication for
In 2019, the average number of procedures performed surgery, primary and revision operations, where surgery
per consultant across all ankle replacements was 6.4 was carried out and how it was funded. Descriptive
dropping to 4.0 in 2020. analyses will be used to look at how the proportion
of any missing PROMs data varies over time, by
In 2020 the Infinity ankle replacement implant looking at trends for each year of data. We will further
dominated at 65.2% of all total ankle replacements, explore geographical variation in patterns of missing
although it was only introduced in 2014. It is reassuring data by hospital trust, and operating surgeon, and
to see that the short term implant survival has improved. use caterpillar plots to visually display hospitals and
surgeons with the most and least amount of missing
Overall revision rates appear to be just less than 9%
PROMs data.
at ten years, although we believe there is incomplete
reporting of conversion to fusion which remains We will present descriptive statistics to look at the
mandatory. Both the Star and the Infinity ankle implant association of patient characteristics according to
are running at revision rates of less than 3% at five completion of PROMs scores. We will do these
years, albeit running into low numbers with longer analyses overall, and then repeat them for individual
follow-up. Other implants are failing at varying, and years of the data, as the influence of patient
some at concerning, rates. characteristics on missing PROMs outcomes and the
quality of data may change over time. Caterpillar plots
Patient Reported Outcome will describe the variation in PROMs outcomes within
Measures (PROMs) and between implant brands and constructs. Initial
analyses will be descriptive about the actual variation
Our annual report includes the failure rates of all the in observed PROMs outcomes, with stratification of
different brands used in hip and knee replacements, analyses by age and gender groups. More formal
however revision surgery is not a complete marker of modelling methods will then be considered, for ‘within
success. A device may well be classified as successful and between group’ variation between implant brands
if it survives for 15 years, but an implanted patient and constructs.
may disagree should they have experienced persistent
pain and disability and around 20% of patients report Having assessed data quality, our aim is to then
persistent pain following joint replacement surgery. compare and scrutinise the differing performances
Analysis of revision rates alone fails to identify these between the implant brands of different prostheses for
patients with persistent pain or disability. Therefore, associated pain and functional outcomes. By selecting
there have been calls for methods to measure pain that a hip or knee brand with a highest improvement in pain
can subsequently be used in conjunction with revision and functional outcome as a reference group, we will
rates for accurately monitoring outcomes in hip and perform statistical analyses to directly compare the
knee replacement surgery. See also the Independent performance of all the stem and cup combinations
Medicines and Medical Devices Safety Review, chaired used in hip replacement and all the knee brands used
by Baroness Julia Cumberlege, published report titled in knee replacement against this reference. This will
“First Do No Harm” (Cumberlege, 2020). demonstrate if any brands are performing poorly in
comparison to the best performing implants, and thus
Patient reported measures of hip and knee pain and enable patients and surgeons to make better informed
function have been collected nationally by NHS Digital decisions about the relative performance, as judged
since 2009 for all patients receiving a primary hip or by patient reported pain and functional outcome of the
knee replacement operation. This information, to date, construct of each brand.
has not been reported in our annual report. We aim
Cumberlege J. First Do No Harm: The report of the Independent Medicines and Medical Devices Safety Review. 2020 Jul 8. https://www.immdsreview.org.uk/Report.html
Concluding acknowledgements Getting It Right First Time (GIRFT)
British Orthopaedic Association (BOA)
The NJR continues to work collaboratively with our
British Hip Society (BHS)
many stakeholders; the most important, of course,
are the patients we serve, and whom we would like British Association for Surgery of the Knee (BASK)
to thank for allowing us to use their data. The NJR British Elbow and Shoulder Society (BESS)
operational collaboration is a huge team effort – this
British Orthopaedic Foot and Ankle Society (BOFAS)
year managed almost exclusively by work performed
virtually. Elaine Young, NJR Director of Operations has European Orthopaedic Research Society (EORS)
demonstrated the great versatility of her leadership and Healthcare Quality Improvement Partnership (HQIP)
her team.
NEC Software Solutions UK Ltd (previously known
Many thanks also to the following without which the as Northgate Public Services UK Ltd)
NJR could not function: University of Bristol
All members of the NJR Steering Committee University of Oxford

Confidentiality Advisory Group (CAG)
Members of the NJR sub-committees:
Association of British HealthTech Industries (ABHI)
Executive
Data Quality We are most grateful to our contractors for their very
valuable input into the NJR Annual Report, and many
Editorial Board
other functions. NEC Software Solutions, University of
Implant Scrutiny Bristol and University of Oxford teams help us refine and
Medical Advisory improve each year. This year’s report is the biggest and
best report yet. We offer our personal thanks to Vicky
Regional Clinical Coordinators
McCormack, Report Project Manager and Deirdra Taylor,
Research Associate Director of Communication and Stakeholder
Surgical Performance Engagement for the NJR, for getting the final report into
shape in the face of challenging circumstances.
Members of the Data Access Review Group
On a personal note, we would particularly like to thank
Members of the NJR Patient Network Laurel Powers-Freeling, Chairman of the NJR. Laurel’s
leadership over the last ten years has seen the NJR grow
Other organisations: in terms of size, quality, stature, and utility. Laurel brought
Medicines and Healthcare products huge insight to the NJR from her many other areas of
Regulatory Agency (MHRA) expertise and her guidance has enabled the organisation
Care Quality Commission (CQC) to grow in ways we would simply not have been able to
develop without those insights. We owe her a great deal
NHS England and Improvement
and wish her every success in her future endeavours,
NHS Digital followed by a long and happy retirement.
Professor Mike Reed Mr Tim Wilton

and
Chairman of the NJR Editorial Board NJR Medical Director
3. Outcomes after
joint replacement
2003 to 2020
3.1 Summary
of data sources,
linkage and
methodology
The main outcome analyses in this report relate Data quality:
to primary and revision joint replacements, unless
otherwise indicated. We included all patients with High quality data are the foundation of any joint
at least one primary joint replacement carried out replacement registry and the National Joint Registry
between 1 April 2003 and 31 December 2020 fully understands and endorses this. From inception,
inclusive, whose records had been submitted to the it was mandatory to record hip and knee arthroplasty
registry before 1 March 2021. procedures for the independent sector but not initially
so in NHS hospitals. It was not until 1 April 2011 that
Information governance and it also became mandatory to enter publicly financed
patient confidentiality: (NHS) procedures into the registry.
Data are collected via a secure web-based data When the NJR was established, the funding model was
entry application and stored and processed in NEC based on a levy system. The manufacturer collected a
Software Solutions’ (NEC) data centre. NEC is ISO small levy for every construct they sold. This practice
27001 and ISO 9001 accredited, and compliant continued from 2003 to 2014 after which the funding
with the NHS’s Data Security and Protection Toolkit. model changed. This levy system generated an
Data linkage to other datasets is approved by the additional source of data from which we could compare
Health Research Authority under Section 251 of sales to uploads into the registry. This process gave a
the NHS Act 2006. Please visit https://www.hra. crude estimate of compliance and for the first four years
nhs.uk/about-us/committees-and-services/ of the registry, compliance could have been improved.
confidentiality-advisory-group/. Post-2008 the compliance rate was in excess of 95%
and on occasion greater than 100% (see Figure 3.D1).
When compliance was over 100%, this was indicative
of the practice of stockpiling prostheses.
Figure 3.D1 Compliance rates from 2003 to 2014.
120
© National Joint Registry 2021
100
80
Percentage
60
40
20
Financial year 2003/04 2004/05 2005/06 2006/07 2007/08 2008/09 2009/10 2010/11 2011/12 2012/13 2013/14
Compliance 43.4 68.3 81.6 80.5 95.2 91.5 114.0 106.6 89.9 89.9 99.6
Comparing procedures to a levy had utility, however procedures uploaded to the registry against a local
it was not sufficiently refined to distinguish within-year hospital’s Patient Administration System (PAS).
compliance and differential-compliance in the upload Records were identified from the local hospital-based
of primary and revision procedures. An additional OPCS4 codes and then matched to records held
comparator was therefore needed to properly assess within the registry, see Figure 3.D2. Records that were
compliance, and the Hospital Episode Statistics (HES) found on the local hospital PAS but not on the registry
service has been used for this purpose for hospitals in were subsequently uploaded bringing compliance
England since 2006. as near to 100% as possible. This procedure could
not be followed if the patient had not given consent
The comparison of data entry onto the registry and to data release. It was expected that neither the
HES data gave a clear indication of the degree to registry nor the local hospital’s PAS system alone
which data might be missing or of any anomalies in could be regarded as a definitive list of hip or knee
data-entry, but does not itself supply or correct the replacements, however, the union of both registry and
missing data. For this reason a formal audit cycle, local hospital data was considered the gold standard
capable of reconciling the two sources of data and from which to calculate voluntary unprompted
allowing their correction was set up using data from compliance at upload. This figure is important for
each NHS hospital’s Patient Administration System healthcare provider institutions as a measure of
(PAS) and each independent hospital’s business compliance with data entry processes but does not
administration system. represent the final data completeness of records
in the registry. It is important to note that nearly all
In 2015 a comprehensive retrospective audit of 149
unmatched procedures identified by the audit were
NHS trusts for procedures uploaded in the 2014/15
subsequently uploaded into the registry.
financial year was initiated. This audit compared
Figure 3.D2 Schematic presentation of NJR data compliance audit.
National Local
Joint Registry Hospital © National Joint Registry 2021
NJR
Data compliance audit
Unmatched
procedures
The audit was expanded to include hip and knee of all hips and knees recorded by the NJR. Since then
procedures performed in the independent sector in the the audit process has been repeated each year.
2015/16 financial year, ensuring complete coverage
Table 3.D1 Percentage data quality audit compliance.

Percentage missing NJR records (%)

Procedure 2014/15 2015/16 2016/17 2017/18 2018/19
Hip primary 4.3 5.4 4.19 4.16 2.38
Hip revision 8.1 11.42 8.74 9.15 5.02
Knee primary 3.5 4.86 3.83 3.41 1.52
Knee revision 8.8 12.45 9.25 8.77 4.79
Note: Percentages for years prior to 2018/19 are pre-audit figures prior to introduction of the automated audit process. Percentages for the 2018/19 audit are as
at 10 August 2021 using the automated process.
The recording of revision procedures in the registry Analysis of data which is missing in either a random
has noticeably improved since the audit has been in (Missing Completely At Random) fashion or random
place. In the most recently completed audit for years within known strata (Missing At Random), e.g. method
2018/19, 97.87% of NJR hip and knee records could of fixation, is known to yield unbiased results. We
be matched to HES and local PAS systems. believe that a coordinated systematic agreement of
individuals across the registry to under-report the
During this last year we have undertaken a national roll failure of a specific implant is exceedingly unlikely.
out of an automated process enabling units to check Nevertheless, we believe if this did happen the
their data quality on a monthly or quarterly basis. This is issue would be identified and corrected by the audit
underway for hip, knee, elbow, ankle and shoulder data process. The low revision rates of either hip or knee
and the pilot suggested that this will rapidly become replacements also makes it exceedingly difficult to
part of normal workflow and greatly reduce the number predict which is likely to fail. Therefore, planning to
of mismatches between registry and hospital data. omit selected primary joint replacements which are
We anticipate that compliance and data accuracy will anticipated to fail within ten years following surgery
exceed 99% when the process is fully embedded. would be unlikely to succeed. Increased centralisation
of revision joint replacement, by specialist revision
Missing data: surgeons, also means there is little motivation to omit
revision which would largely have been primary cases
The effect of missing data on the statistical analysis of
of another surgeon or another unit.
data is well documented. Data which is systematically
missing (Missing Not at Random) has the potential We believe that missing data within the registry can
to induce bias i.e. to distort the truth. This is why be considered missing completely at random. We
compliance of reporting data to the registry by a propose that this missing data mechanism will ensure
specific consultant or unit is essential to the quality that the quality assurance process of prostheses
assurance process of consultants and units. entered into the registry, consultant and units is
statistically valid.
Patient level data linkage: be seen at the beginning of each sub-section of

each type of joint replacement. Cases from Northern
Documentation of implant survivorship and mortality Ireland and the States of Guernsey were also excluded
requires linkage of person-level identifiers in order to because of as yet unresolved issues around tracing
identify primary and revision procedures and mortality mortality; and cases from the Isle of Man were also
events for the same individual. excluded due to the inability to audit them against
local hospital data. Patients with longer follow-up may
Starting with a total of 3,152,913 NJR sourced be less representative of the whole cohort of patients
records, 6.7% were excluded because no suitable undergoing primary joint replacement than those
person-level identifier was found (see Figure 3.D3). patients with shorter follow-up, due to difficulties with
Full details of the inclusion and exclusion criteria can data linkage and differential rates of reporting over time.
Figure 3.D3 Initial numbers of procedures for analysis.
3,152,913 210,642
No (Consent | Tracing | ID)
procedures
2,942,271
Consenting & Traced & ID

2,941,968
procedures with
concordant patient &
date data
2,895,368
procedures with a
consistent operative
pattern
HIPS: KNEES: ANKLES: ELBOWS: SHOULDERS:

1,251,164 1,357,077 7,084 5,043 50,255
primaries primaries primaries primaries primaries
129,308 87,535 869 1,417 5,616
revisions revisions revisions revisions revisions
Linkage between primaries and Analytical methods and terminology
any associated revisions The NJR Annual Report uses a variety of statistical
(the ‘linked files’): methods to reflect the diversity and range of
performance within joint replacement. Analyses are
A total of 2,670,623 linked and analysable primary joint
tailored to ensure results are reported in units that
replacements have been recorded by the NJR, i.e. hip,
can be easily interpreted. Here we define important
knee, ankle, shoulder or elbow. Implant survivorship is
concepts which underpin the analyses in the
first described with respect to the lifetime of the primary
following sections.
joint only. In sections 3.2 and 3.3, we also provide an
overview of further revisions following the first hip or All cause / all construct revision
knee revision procedure.
All cause revision is used as the primary outcome in
As in previous years, the unit of observation for all the majority of analyses due to the difficulties in defining
sets of survivorship analysis has been taken as the cause-specific failure i.e. several indications may have
individual primary joint replacement. A patient with left been given for a particular revision. In addition, we
and right replacements of a particular type, therefore, consider the construct as a single entity, for example,
will have two entries, and an assumption is made in hips we do not differentiate between stem and
that the survivorship of a replacement on one side is acetabular failure as it is sometimes difficult to identify
independent of the other. In practice, this would be which prosthetic element failed first or is causally
difficult to validate, particularly given that some patients responsible for the failure. It is incorrect to assume that
will have had primary replacements of other joints the failure of implants that make up a construct are
that were not recorded in the registry. Established independent of each other. In knees, we similarly do
risk factors, such as age, are recorded at the time of not differentiate between failure of components within
primary operation and will therefore be different for the tibia, femur or patella. Secondary patella resurfacing
the two procedures unless the two operations are after a total knee replacement is considered a revision.
performed on the same date. In shoulders, elbows and ankles we take the same
approach and do not differentiate between the failure
A revision is defined as any operation where one or of different components within the joint. Conversions
more components are added to, removed from or of one type of shoulder replacement to another are
modified in a joint replacement, or if a Debridement considered a revision.
And Implant Retention (DAIR) with or without modular
exchange is performed. Capturing DAIR with or without Debridement And Implant Retention
modular exchange commenced with the introduction of Debridement And Implant Retention (DAIR) without
MDSv7. Prior to this DAIR with modular exchange was modular exchange has been included in the registry
included as a single-stage revision but DAIR without data as of MDSv7 (June 2018). DAIRs with modular
modular exchange was not captured. Within the annual exchange should have been collected (as a type of
report, each of these procedure types is included in the single-stage revision) from inception and their reporting
analyses as a revision episode. This is distinct from the in hips, knees, shoulders and elbows, along with all
analyses in the surgeon, unit, and implant performance other procedures captured by the NJR, has been
workstreams where DAIR without modular exchange is mandatory since 1 April 2011. Before MDSv7, DAIRs
not currently included as a revision outcome. with modular exchange were considered to be a
revision in hip, knee, shoulder and elbow but not ankle
replacements. In MDSv7, all joint types are treated the implanted and in future reports these will be reported
same and a DAIR with modular exchange is considered as a new category, although the numbers are likely to
to be a revision in all recorded joint replacements. remain too small for meaningful analysis for a number
of years. Three bearing materials being listed indicates
Terminology note: Hip replacements the use of dual-mobility bearing devices. The size of
There are four distinctive categories reflected in the the femoral head or inner diameter of a component is
analysis of data collected in the registry and these expressed in millimetres.
are: 1) the type of hip replacement i.e. total hip
Terminology note: Knee replacements
replacements (THR) and hip resurfacings (the NJR
does not currently collect data on hip hemiarthroplasty); Knee replacements within the registry are principally
2) the fixation of the replacement i.e. cemented, defined by the number and type of compartments
uncemented, hybrid and reverse hybrid; 3) the replaced, the fixation of the components (cemented,
bearing surfaces of the hip replacement; and 4) the uncemented or hybrid), level of constraint, the mobility
size of femoral head/internal diameter of the of the bearing, whether the implants are of a modular
acetabular bearing. design and the presence or absence of a patella in the
primary knee replacement.
Cemented constructs are fixed using bone
cement in both the femoral stem and acetabulum. The knee is made up of three compartments:
Uncemented constructs rely on press fit and osseous medial, lateral and patellofemoral. When a total knee
integration within the femur and acetabulum that replacement (TKR) is implanted, the medial and
may be supplemented (e.g. by screw fixation). Hybrid lateral compartments are always replaced, and the
constructs contain a cemented femoral stem and an patella may be resurfaced. If a single compartment
uncemented acetabulum. Reverse hybrid constructs is replaced then the term unicompartmental is
contain an uncemented femoral stem and a cemented applied to the procedure (UKR). The medial, lateral
acetabulum. By convention, the bearing material of the or patellofemoral compartments can all be replaced
femoral head is listed before the acetabulum. independently, if clinically appropriate. Medial and lateral
Currently, the eight main categories of bearing surfaces unicompartmental knee replacements are also referred
for hip replacements are ceramic-on-ceramic (CoC), to as medial or lateral unicondylar knee replacements.
ceramic-on-metal (CoM), ceramic-on-polyethylene We also use the term multicompartmental knee
(CoP), metal-on-metal (MoM), metal-on-polyethylene replacement to indicate the combination of more than
(MoP), metal-on-polyethylene-on-metal (MoPoM), one unicompartmental knee replacement.
ceramic-on-polyethylene-on-metal (CoPoM), and
resurfacing procedures. Knee replacements are also characterised by their
level of constraint (stabilisation). For example, there is
The metal-on-metal group in this section refers to variation in the constraint of the tibial insert’s articulation
patients with a stemmed prosthesis (THR) and metal with the femoral component depending on whether
bearing surfaces (a monobloc metal acetabular cup the posterior cruciate ligament is preserved (cruciate
or a metal acetabular cup with a metal liner). Although retaining; CR) or sacrificed (posterior stabilised; PS)
they have metal-on-metal bearing surfaces, resurfacing at the time of surgery. Additional constraint may be
procedures, which have a surface replacement femoral necessary to allow the implant to deal with additional
prosthesis combined with a metal acetabular cup, are ligament deficiency or bone loss (where constrained
treated as a separate category. Ceramic-on-ceramic condylar (CCK) or hinged knee implants would be used)
and metal-on-polyethylene resurfacings are now being in a primary or revision procedure.
In modular tibial components, the tibial insert may be Terminology note: Shoulder replacements
mobile or remain in a fixed position on the tibial tray.
Shoulder replacements within the registry are principally
This also applies to medial and lateral unicompartmental
defined by the type and sub-type of replacement.
knees. Many brands of total knee implant exist in
The four main types of replacement are 1) proximal
fixed and mobile forms with options for either CR or
humeral hemiarthroplasty, 2) conventional total
PS constraint. Tibial elements may or may not be of
shoulder replacement, 3) reverse polarity total shoulder
modular design. Modularity allows some degree of
replacement and 4) interpositional arthroplasty. There
patient-specific customisation. For example, modular
are three main sub-types based on variations on the
tibial components are typically composed of a metal
humeral side of the joint. These include 1) resurfacing
tibial tray and a polyethylene insert which may vary
i.e. putting a new metal surface over the existing
in thickness. Non-modular tibial components consist
humeral head, 2) stemless i.e. removing the humeral
of an all-polyethylene tibial component (monobloc
head and putting on a new head with an anchoring
polyethylene tibia) available in different thicknesses.
device which does not project beyond the metaphysis
We now distinguish between medial and lateral of the proximal humerus, and 3) stemmed i.e. replacing
unicondylar knee replacements during the data the humeral head and utilising an anchoring device
collection process; however this was not so in earlier which projects into the diaphysis of the humerus.
versions of the minimum dataset form (MDS) i.e. those
Descriptive statistics
prior to MDSv7.
In simple cases we tend to report simple descriptive
In addition, we now report multicompartmental knee statistics including: numbers (n), frequencies (N=),
replacements which may include unicondylar and percentages (%), minimums (min), maximums (max),
patellofemoral or two unicondylar replacements. interquartile ranges (IQR) (25th centile, 75th centile),
means (SD) and medians (50th centile) of the data.
With regard to the use of the word ‘constraint’ here,
for brevity, total knee replacements are termed Survival analysis methods
unconstrained (instead of posterior cruciate-retaining)
In more complex analyses that focus on either implant
or posterior-stabilised (instead of posterior
failure (denoted revision), recurrent implant failure (re-
cruciate-stabilised).
revision) or mortality we use ‘survival analysis methods’
We assume the absence of a patella in the upload of which are also known as ‘time to event’ methods.
knee components is indicative that the patella has not
Survival analysis methods are necessary in joint
been resurfaced.
replacement data due to a process known as
Terminology note: Ankle replacements ‘censoring’. There are two forms of censoring which are
important to consider in joint replacement registry data:
Ankle replacements recorded within the registry are administrative censoring and censoring due to events,
principally uncemented devices. However, in terms such as death.
of fixation we now report the presence or absence of
cement used within the ankle construct. The presence Administrative censoring creates differential amounts of
of cement is defined by the inclusion of cement product follow-up time, i.e. patients from 2003 will have been
details within the prosthesis upload. followed up for more than 17 years, whilst patient data
collected last year will have one year of follow-up or
less. Survival analyses methods enable us to include in the numbers at risk because of censoring do not
all patients in one analysis without being concerned themselves cause a step change but if the numbers
if patients have one day, one year or one decade of at risk become low, when an event does occur, the
observed follow-up time; these methods automatically stair-step might appear quite dramatic. Whenever
adjust analyses for the amount of follow-up time. possible, the numbers at risk at each time point have
been included in the figures, allowing the reader to
In the case of analyses which estimate implant failure, more appropriately interpret the data given the number
death events are also censored, specifically they are of constructs at risk. We highlight in blue italics all
considered non-informative censoring events. This estimates where there are less than 250 prosthesis
assumes that death is unrelated to a failing implant, constructs at risk or remaining at risk at that particular
and can be safely ignored whilst estimating implant time point. The Kaplan-Meier estimates shown are
failure (revision). See Sayers et al. 2018 Acta technically 1 minus the Kaplan-Meier estimate multiplied
Orthopaedica, 89:3, 256-258, for an extensive by 100, therefore they estimate the cumulative
discussion on this problem. percentage probability of construct failure.
The survival tables in this report show ‘Kaplan-Meier’ In the case of revisions, no attempt has been made
estimates of the cumulative chance (probability) of to adjust for the risk of death, as analyses attempt
failure (revision) or death, at different times from the to estimate the underlying implant failure rate in
primary operation. In the joint replacement literature the absence of death, see Sayers et al. 2018 Acta
they are often referred to as KM or simply survival Orthopaedica, 89:3, 256-258 for an extensive
estimates. We additionally show 95% Confidence discussion on competing risks. Briefly, the Kaplan-
Intervals for each estimate (95% CI). Confidence Meier estimator estimates the probability of implant
intervals illustrate the uncertainty around the estimate, failure (revision) assuming the patient is still alive.
with wide confidence intervals indicating greater
uncertainty than narrow ones. Strictly they are Prosthesis Time Incidence Rates
interpreted in the context of repeated sampling i.e. if
Prosthesis Time Incidence Rates (PTIR) are used to
the data were collected in repeated samples we would
describe the incidence (the rate of new events) of
expect 95% CIs generated to contain the true estimate
specific modes of failure in joint replacement. The PTIR
in 95% of samples. However, confidence intervals
expresses the number of revisions divided by the total
are strongly influenced by the numbers of prosthesis
of the individual prosthesis-years at risk. Figures here
constructs at risk and can become unreliable when
show the numbers of revisions per 1,000 years at
the numbers at risk become low. In tables, including
risk. PTIR in other areas of research are often known
risk tables within figures, we highlight in blue italics all
as ‘person-time’ incident rates, however, in joint
estimates where there are less than 250 prosthesis
replacement registries the base unit of analysis is the
constructs at risk, or remaining at risk, at that particular
‘prosthesis construct’.
time point.
Note: This method is only appropriate if the hazard
Kaplan-Meier estimates can also be displayed
rate (the rate at which revisions occur in the unrevised
graphically using a connected line plot. Figures are
cases) remains constant across the follow-up period.
joined using a ‘stair-step’ function. Each ‘stair’ is flat,
The latter is further explored by sub-dividing the time
reflecting the constant nature of the estimate between
interval from the primary operation into smaller intervals
the events of interest. When a new event occurs the
and calculating PTIRs for each smaller interval.
survival estimate changes, creating a ‘step’. Changes
3.2 Outcomes after
hip replacement
National Joint Registry | 18th Annual Report | Hips
3.2.1 Overview of primary hip December 2020), 250,278 primary hip procedures
(representing 20.0% of the current registry volume)
replacement surgery were performed by 2,164 consultant surgeons
In this section we address revision and mortality working across 424 units.
outcomes for all primary hip operations performed
Looking at caseload over this three-year period, the
between 1 April 2003 and 31 December 2020.
median number of primary procedures per consultant
Patients operated on at the commencement of
surgeon was 60 (interquartile range (IQR) 4 to 183)
the registry therefore had a potential 17.75 years
and the median number of procedures per unit was
of follow-up. This year, follow-up is reported at a
525 (IQR 263 to 814). A proportion of surgeons will
maximum of 17 years in the tables and figures,
have commenced practice as a consultant during this
although beyond 15 years the numbers at risk are
period, some may have retired, and some surgeons
particularly low in some categories.
may have periods of surgical inactivity within the time
Figure 3.H1 (a) (page 46) describes the data cleaning of coverage of the registry, therefore their apparent
applied to produce the total of 1,251,164 hip caseload would be lower.
procedures included in the analyses presented in
The majority of primary hip procedures were carried
this section.
out on women (females 59.9%: males 40.1%). The
Over the lifetime of the registry, the 1,251,164 primary median age at primary operation was 69 (IQR 61 to
hip replacement procedures contributing to our 76) years. Osteoarthritis was given as a documented
revision analyses were carried out by a total of 3,821 indication for surgery in 1,142,684 cases (91.3%
unique consultant surgeons working across 478 of the cohort) and was the sole indication given in
units. Over the last three years (1 January 2018 to 31 1,102,840 (88.1%) primary hip replacements.
Figure 3.H1 (a) Hip cohort flow diagram.
Hip procedures recorded by the NJR

N=1,514,612
*Reoperation procedures N=1,354
*Non−consenting procedures N=60,050
*Non−traced procedures N=47,836
*Invalid IDs N=1
*Unknown procedures N=1
Consenting / Traced / With valid IDs

N=1,405,850
*Procedures prior to April 2003 N=48
*Patients who died before their operation date N=34
*Procedures with a listed age <0 or >100 years N=51
*Patient procedures ≥110 years old
at administrative censoring date N=12
Procedures with concordant date information

N=1,405,716
*No gender recorded N=5
*No side recorded N=0
Procedures with concordant patient information

N=1,405,711
Northern Ireland N=17,954

Isle of Man N=679
States of Guernsey N=0
Unknown N=1
English and Welsh procedures

N=1,387,077
Duplicate primary procedures based on:
NHS No. / Date / Side / Age at op.
/ Gender / ASA grade / Procedure type
/ Prostheses used / Indications / Unit N=1,039
Duplicate same day revision procedures based on:
NHS No. / Date / Side / Procedure type N=50
Unique procedures
N=1,385,988
Procedures (2,626 hips) with
an inconsistent operative pattern N=5,516
Procedures (1,324,349 hips)

with a consistent operative pattern
N=1,380,472
*All revision procedures N=129,308

*Of which, hip procedures where the first recorded
procedure in a sequence is a revision N=84,275
Primary procedures
(Revision analyses)
N=1,251,164
Bilateral procedures (same day) N=5,449
Ipsilateral procedures
(Mortality analyses)
N=1,245,715 * Reasons not necessarily mutually exclusive
Table 3.H1 Number and percentage of primary hip replacements by fixation and bearing.
Percentage of each
Number of bearing type used
Bearing surface within primary hip within each method Percentage of all
Fixation fixation group operations of fixation primary hip operations
All cases 1,251,164 100.0
All cemented 391,414 31.3
MoP 338,744 86.5 27.1
MoM 407 0.1 <0.1
CoP 49,677 12.7 4.0
MoPoM 2,340 0.6 0.2
Others 246 0.1 <0.1
All uncemented 465,982 37.2
MoP 181,446 38.9 14.5
MoM 29,028 6.2 2.3
CoP 118,388 25.4 9.5
CoC 133,721 28.7 10.7

CoM 2,153 0.5 0.2
MoPoM 725 0.2 0.1
CoPoM 406 0.1 <0.1
Others 115 <0.1 <0.1
All hybrid 284,326 22.7
MoP 159,609 56.1 12.8
MoM 2,722 1.0 0.2
CoP 90,769 31.9 7.3
CoC 26,961 9.5 2.2
MoPoM 3,236 1.1 0.3
CoPoM 862 0.3 0.1
Others 166 0.1 <0.1
All reverse hybrid 32,596 2.6
MoP 22,231 68.2 1.8
CoP 10,141 31.1 0.8
Others 224 0.7 <0.1
All resurfacing 40,081 3.2
MoM 39,883 99.5 3.2
Others 198 0.5 <0.1
Unclassified 36,765 2.9
Table 3.H1 shows the breakdown of cases by the either as dual mobility, to contrast to standard unipolar
method of fixation and within each fixation sub-group, bearings, or where numbers allow, are categorised
by bearing surfaces. Bearing surface combinations by the material of each part of the bearing surface
are reported as a separate group where there were (e.g. metal-on-polyethylene-on-metal (MoPoM) and
more than 250 cases. The most commonly used ceramic-on-polyethylene-on-metal (CoPoM)). The
operation type overall remains as cemented metal-on- numbers of other combinations of dual mobility (such
polyethylene (86.5% of all cemented primaries, 27.1% as ceramic-on-polyethylene-on-ceramic (CoPoC))
of all primaries). Dual mobility bearings are described were too small to include as separate groups this year.
Figure 3.H1 (b) Frequency of primary hip replacements within elective cases stratified by procedure
type. Consultants have been placed in groups by the volume of cases they undertake per annum. Each
colour represents total volume of cases undertaken by all the consultants in that grouping.
Unipolar THR
100,000
80,000
60,000
40,000
20,000
0 2003 2005 2007 2009 2011 2013 2015 2017 2019
2004 2006 2008 2010 2012 2014 2016 2018 2020
Resurfacing THR
6,000
4,000
2,000
Frequency (N=)
0 2003 2005 2007 2009 2011 2013 2015 2017 2019

2004 2006 2008 2010 2012 2014 2016 2018 2020
Resurfacing THR year>2011
1,000
800
600
400
200
0 2003 2005 2007 2009 2011 2013 2015 2017 2019
2004 2006 2008 2010 2012 2014 2016 2018 2020
Dual Mobility THR
1,000
800
600
400
200
0 2003 2005 2007 2009 2011 2013 2015 2017 2019
2004 2006 2008 2010 2012 2014 2016 2018 2020
N =Procedures per year and by type, summed over elective and acute replacements
1≤N≤2 3≤N≤4 5≤N≤6 7≤N≤12 13≤N≤24 25≤N≤48 49≤N≤96 97≥N≤192 ≥193
Graphs by confirmed procedure type
Figure 3.H1 (c) Frequency of primary hip replacements within acute trauma cases stratified by
procedure type. Consultants have been placed in groups by the volume of cases they undertake per
annum. Each colour represents total volume of cases undertaken by all the consultants in that grouping.
Unipolar THR
5,000
4,000
3,000
2,000

1,000
Frequency (N=)
0 2003 2005 2007 2009 2011 2013 2015 2017 2019

2004 2006 2008 2010 2012 2014 2016 2018 2020
Dual Mobility THR
300
200
100
0 2003 2005 2007 2009 2011 2013 2015 2017 2019

2004 2006 2008 2010 2012 2014 2016 2018 2020
1≤N≤2 3≤N≤4 5≤N≤6 7≤N≤12 13≤N≤24 25≤N≤48 49≤N≤96 97≥N≤192 ≥193
Figure 3.H1 (b) and Figure 3.H1 (c) show the yearly Figure 3.H1 (b) also shows that after declining
number of primary total hip replacements performed substantially in popularity, resurfacing has only declined
for elective and acute trauma indications respectively. marginally in the past five years. The procedure has
Elective procedures have been stratified by unipolar, declined more among surgeons who undertook
resurfacing and dual mobility total hip replacements. low volumes of resurfacing. In 2020 over half of the
Acute trauma procedures have been stratified by resurfacing procedures were performed by consultants
unipolar and dual mobility total hip replacements, who used it in more than 25 cases per year.
please note the difference in scale of the y-axis
between each sub-plot. Figure 3.H1 (b) and Figure 3.H1 (c) also illustrate the
emerging use of dual mobility THR in the elective
Each bar is further stratified by the volume of and acute trauma setting. Prior to 2013 dual mobility
procedures that the consultant conducted in that THR was relatively rare but since 2013 its use
year across both elective and acute trauma settings has increased in both settings, and it is now more
i.e. if a surgeon performed 25 elective unipolar THR common than hip resurfacing. Over half of dual
procedures and 25 acute trauma unipolar elective mobility operations are performed by consultants who
procedures their annual total volume would be 50 conduct seven or more replacements per year.
procedures. Those 50 procedures would contribute
to the black sub-division in both elective and acute
trauma figures.
Figure 3.H1 (b) shows the annual rates of elective

unipolar THR increasing, (with the exception of 2020
due to the COVID-19 pandemic), with the majority
of additional procedures contributed by higher
volume surgeons i.e. those performing more than 49
procedures a year. A similar result is also observed
in the acute trauma setting with a rapid expansion of
unipolar THRs being recorded in the registry after 2011.
Table 3.H2 Percentage of primary hip replacements by fixation, bearing and calendar year.
2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020
Fixation/ n= n= n= n= n= n= n= n= n= n= n= n= n= n= n= n= n=
bearing 42,578 40,662 48,511 60,895 67,434 68,599 71,119 74,076 78,282 80,438 87,682 89,840 94,346 96,424 96,771 98,649 54,858
All cemented 53.6 46.0 40.3 37.4 31.9 30.0 29.5 30.2 31.8 32.1 31.1 30.1 28.5 27.4 26.9 25.7 22.4
Cemented by bearing surface:
MoP 50.5 43.0 37.4 34.8 29.2 27.3 26.4 26.7 27.8 27.7 26.3 25.1 23.4 22.0 21.7 20.3 17.3
MoM 0.1 0.1 0.2 0.2 0.1 <0.1 <0.1 <0.1 0 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 0
CoP 3.0 2.9 2.8 2.4 2.6 2.7 3.1 3.4 3.9 4.3 4.5 4.6 4.7 4.9 4.9 5.1 4.8
MoPoM 0 0 <0.1 <0.1 <0.1 <0.1 0.1 0.1 0.1 0.1 0.3 0.4 0.4 0.4 0.3 0.3 0.3
Others 0 0 0 0 0 <0.1 0 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 0.1 0.1 0.1
All uncemented 18.3 24.2 28.3 31.5 37.3 40.8 43.2 42.8 44.1 41.9 40.3 39.0 38.2 37.5 36.5 35.1 34.8
Uncemented by bearing surface:
MoP 7.5 9.4 9.6 10.1 12.3 14.4 16.0 16.5 17.5 17.2 16.8 16.2 15.9 15.6 15.2 13.6 12.5
MoM 1.9 5.4 8.3 10.4 11.1 7.9 3.2 0.4 0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1
CoP 5.0 5.1 4.5 4.0 3.8 4.5 5.4 5.9 7.2 8.2 9.5 11.4 12.5 14.1 14.9 15.9 17.1
CoC 3.9 4.3 5.8 6.9 9.7 13.1 17.4 19.5 19.1 16.3 14.0 11.4 9.7 7.6 6.2 5.3 4.8
CoM <0.1 <0.1 <0.1 0.1 0.4 0.9 1.0 0.5 0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1
MoPoM <0.1 <0.1 0 0 0 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 0.1 0.1 0.1 0.1 0.2 0.2
CoPoM 0 0 0 0 0 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 0.1 0.1 0.2
Others <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 0
All hybrid 12.7 13.9 15.1 14.8 14.7 15.4 15.8 16.7 17.4 19.9 22.8 25.3 27.7 29.8 31.6 34.8 37.9
Hybrid by bearing surface:
MoP 8.8 9.4 9.9 9.9 9.9 10.4 10.7 11.3 11.4 11.9 13.1 14.0 14.9 15.5 15.2 16.5 16.2
MoM 0.7 0.6 0.7 0.8 0.7 0.4 0.2 <0.1 <0.1 <0.1 <0.1 <0.1 0.1 0.1 0.1 0.1 <0.1
CoP 1.5 1.2 1.3 1.0 1.3 1.8 1.9 2.2 3.1 5.0 7.0 8.9 10.7 12.3 14.5 16.2 19.4
CoC 1.7 2.7 3.2 2.9 2.7 2.9 3.0 3.1 2.9 2.7 2.4 2.1 1.6 1.4 1.1 0.9 0.7
MoPoM 0 <0.1 0 0 0 <0.1 <0.1 <0.1 0.1 0.1 0.2 0.3 0.4 0.5 0.6 0.8 1.0
CoPoM 0 0 0 0 0 0 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 0.1 0.2 0.3 0.4
Others <0.1 0 0 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1
Note: Percentages calculated as percentage of total yearly operations.
www.njrcentre.org.uk
Note: A zero represents no procedures of this bearing type.
51
52
Table 3.H2 (continued)
2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020
Fixation/ n= n= n= n= n= n= n= n= n= n= n= n= n= n= n= n= n=
bearing 42,578 40,662 48,511 60,895 67,434 68,599 71,119 74,076 78,282 80,438 87,682 89,840 94,346 96,424 96,771 98,649 54,858
All reverse
0.7 0.9 1.0 1.6 2.4 2.6 2.7 3.1 3.1 3.0 3.1 3.2 3.2 3.2 2.9 2.3 2.2
hybrid
Reverse hybrid by bearing surface:
MoP 0.5 0.6 0.8 1.0 1.7 1.8 1.9 2.1 2.0 2.0 2.0 2.1 2.2 2.3 2.1 1.6 1.5
CoP 0.2 0.2 0.2 0.6 0.7 0.8 0.8 0.9 1.1 1.0 1.1 1.0 1.0 0.9 0.8 0.7 0.7
Others <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1
All
9.8 10.3 10.3 9.7 8.3 6.1 3.7 2.4 1.4 1.1 0.9 0.9 0.7 0.7 0.6 0.6 0.9
resurfacing
Resurfacing by bearing surface:
MoM 9.8 10.3 10.3 9.7 8.3 6.1 3.7 2.4 1.4 1.1 0.9 0.9 0.7 0.6 0.5 0.6 0.9
Others 0 <0.1 <0.1 <0.1 0 0 0 0 0 0 <0.1 0 <0.1 0.1 0.1 <0.1 0.1
Unclassified 4.9 4.7 4.9 5.0 5.4 5.0 5.1 4.7 2.3 2.0 1.8 1.6 1.6 1.4 1.5 1.4 1.8
All 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100
Note: Percentages calculated as percentage of total yearly operations.

Table 3.H2 shows the annual rates by fixation but since then has steadily declined to 35% over
and bearing groups for each year for primary hip the last eight years. Figure 3.H2 (a) illustrates the
replacements. The proportion of all hips that are temporal changes in fixation and type of primary hip
cemented has nearly halved between 2006 and replacements. Figure 3.H2 (b) overleaf shows dual
2020. The percentage of hybrid implants used has mobility bearings as a separate group to illustrate their
gone up by over 2.5 times over the same period. The steadily increasing use, which has been most marked
percentage of uncemented implants used increased in the hybrid fixation group (see Table 3.H2).
from 18% to 44% in the first nine years of the registry,
Figure 3.H2 (a) Fixation and type by year of primary hip replacement.
70
60
Percentage of primaries
50

40
30
20
10
0
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020
Year of primary
Cemented Uncemented Hybrid Reverse hybrid Resurfacing
Figure 3.H2 (b) Unipolar THR fixation and main bearing type by year of primary hip replacement.
70
60
50
40
30
20
10
0
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020
Year of primary
Cemented Uncemented Hybrid Reverse hybrid Resurfacing Dual mobility
Figure 3.H3 (a) Cemented primary hip replacement bearing surface by year.
100
90
Percentage of cemented primaries
80
70

60
50
40
30
20
10
0
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020
Year of primary
MoP CoP MoPoM Others
Figures 3.H3 (a) to (d) illustrate the temporal changes

in the bearing surface combinations used by the type
of total hip replacement fixation. Groups that contain
more than 500 procedures are plotted separately.
Since 2012 there has been a marked increase in
the use of ceramic-on-polyethylene bearings and a
corresponding decrease in the use of ceramic-on-
ceramic bearings. The greatest variation in bearing use
is noted in the uncemented fixation group.
Figure 3.H3 (b) Uncemented primary hip replacement bearing surface by year.
50
Percentage of uncemented primaries
40
30
20
10
0
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020
Year of primary
MoP MoM CoP CoC CoM MoPoM Others
Figure 3.H3 (c) Hybrid primary hip replacement bearing surface by year.
70
60
Percentage of hybrid primaries
50

40
30
20
10
0
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020
Year of primary
MoP MoM CoP CoC MoPoM CoPoM Others
Figure 3.H3 (d) Reverse hybrid primary hip replacement bearing surface by year.
80
Percentage of reverse hybrid primaries
70
60
50
40
30
20
10
0
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020
Year of primary
MoP CoP Others
Figure 3.H3 (e) Trends in fixation, bearing and head size in primary unipolar total hip replacement
by year.
22.25 mm 26 mm 28 mm 32 mm 36 mm
30 30 30 30 30
Cemented
20 20 20 20 20
10 10 10 10 10
0 0 0 0 0
03
07
11
15
19
03
07
11
15
19
03
07
11
15
19
03
07
11
15
19
03
07
11
15
19
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
Percentage of all primaries

30 30 30 30 30
Uncemented
20 20 20 20 20
10 10 10 10 10
0 0 0 0 0
03
07
11
15
19
03
07
11
15
19
03
07
11
15
19
03
07
11
11
15
19
03
07
15
19
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
30 30 30 30 30
20 20 20 20 20
Hybrid
10 10 10 10 10
0 0 0 0 0
03
07
11
15
19
03
07
11
15
19
03
07
11
15
19
03
07
11
15
19
03
07
11
15
19
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
Year of primary
MoP CoC CoP MoM
Note: Only combinations with ≥2% use in any year are plotted.
Figure 3.H3 (e) illustrates the temporal changes In 2020, as in 2019, the three most common head
in common head sizes, by method of fixation sizes are 32mm (1st), 36mm (2nd) and 28mm (3rd),
and bearing type in primary unipolar total hip with 22.25mm and 26mm rarely being used. The
replacement. In 2003, the vast majority of hip use of ceramic-on-ceramic bearings across all head
replacements utilised heads of 28mm or smaller, sizes, but most notably 36mm, has declined since
across all fixation methods. Since 2003, a 2011. This decline, conversely, corresponds with an
progressive shift away from small (22.25mm or increase in ceramic-on-polyethylene bearings with
26mm) heads in cemented hip replacements to 32mm heads. The choice of bearing, head size and
larger head sizes (>28mm) with alternative fixation fixation method is much more heterogeneous in 2020
methods (uncemented or hybrid) has been observed. compared to 2003.
Table 3.H3 provides a breakdown by fixation type receiving metal-on-polyethylene-on-metal dual
and bearing surface, describing the age and gender mobility bearings tended to be older than those in
profile of recipients of primary hip replacements. the other groups. Those receiving resurfacings were
Patients receiving resurfacing and ceramic-on- more likely to be younger men.
ceramic bearings tended to be younger and those
Table 3.H3 Age at primary hip replacement by fixation and bearing.
By bearing surface Age (years) Percentage

Fixation within fixation group N Median (IQR*) Mean (SD) males (%)
All cases 1,251,164 69 (61 to 76) 68.1 (11.4) 40.1
All cemented 391,414 74 (68 to 79) 73.1 (9.1) 33.5
Cemented and
MoP 338,744 75 (69 to 80) 74.3 (8.2) 32.8
MoM 407 72 (65 to 78) 71.1 (9.5) 33.7
CoP 49,677 65 (59 to 71) 64.6 (10.4) 37.9
MoPoM 2,340 77 (70 to 83) 75.5 (10.7) 30.9
Others 246 75 (66 to 83) 72.5 (12.9) 29.7
All uncemented 465,982 65 (58 to 72) 64.4 (11.3) 45.0
Uncemented and
MoP 181,446 71 (64 to 76) 69.9 (9.5) 41.5
MoM 29,028 63 (57 to 70) 63.0 (11.1) 50.8
CoP 118,388 64 (57 to 70) 62.9 (10.1) 46.3
CoC 133,721 60 (52 to 66) 58.6 (11.3) 47.2

CoM 2,153 63 (56 to 69) 62.0 (10.6) 42.1
MoPoM 725 71 (61 to 79) 69.0 (13.4) 38.9
CoPoM 406 59 (52 to 69) 60.4 (13.0) 59.1
Others 115 62 (52 to 71) 61.0 (13.7) 45.2
All hybrid 284,326 70 (63 to 77) 69.1 (10.9) 37.3
Hybrid and
MoP 159,609 74 (68 to 79) 73.2 (8.7) 34.8
MoM 2,722 65 (57 to 74) 64.6 (12.4) 46.4
CoP 90,769 66 (59 to 72) 65.0 (10.6) 40.5
CoC 26,961 60 (53 to 66) 59.0 (11.3) 40.8
MoPoM 3,236 76 (68 to 82) 73.8 (11.2) 33.3
CoPoM 862 69 (59 to 77) 67.4 (12.9) 46.1
Others 166 67 (58 to 75) 65.8 (12.8) 48.8
All reverse hybrid 32,596 71 (64 to 77) 69.7 (9.8) 37.0
Reverse hybrid and
MoP 22,231 73 (68 to 78) 72.8 (8.0) 35.7
CoP 10,141 64 (58 to 69) 63.0 (9.7) 40.0
Others 224 72 (57 to 81) 68.0 (15.8) 30.8
All resurfacing 40,081 55 (48 to 60) 53.9 (9.1) 73.4
Resurfacing and
MoM 39,883 55 (48 to 60) 53.9 (9.1) 73.5
Others 198 54 (48 to 60) 53.4 (10.8) 55.1
Unclassified 36,765 69 (61 to 77) 68.0 (12.5) 38.5
*IQR=interquartile range.
Table 3.H4 Primary hip replacement patient demographics.

Males Females All
N (%) N (%) N (%)
Total 502,239 748,925 1,251,164

ASA 1 90,190 (18.0) 104,495 (14.0) 194,685 (15.6)
ASA 2 326,480 (65.0) 519,514 (69.4) 845,994 (67.6)
ASA 3 82,248 (16.4) 120,953 (16.2) 203,201 (16.2)
ASA 4 3,258 (0.6) 3,877 (0.5) 7,135 (0.6)
ASA 5 63 (<0.1) 86 (<0.1) 149 (<0.1)
Osteoarthritis as
the sole reason for 450,019 (89.6) 652,821 (87.2) 1,102,840 (88.1)
primary
Osteoarthritis as a
465,099 (92.6) 677,585 (90.5) 1,142,684 (91.3)
reason for primary
Mean (SD) Median (IQR) Mean (SD) Median (IQR) Mean (SD) Median (IQR)
Age
66.5 (11.6) 68 (59 to 75) 69.2 (11.2) 70 (63 to 77) 68.1 (11.4) 69 (61 to 76)
Table 3.H4 shows the American Society of grade greater than ASA 3 undergo a primary hip
Anesthesiologists (ASA) grade and indication for replacement. The majority of cases are performed for
primary hip replacement by gender. A greater osteoarthritis. A total of 1,102,840 (88.1%) primary
number of females than males undergo primary hip hip replacements have been recorded in the registry
replacement and two-thirds of patients are ASA where the sole indication was osteoarthritis.
grade 2. Only a small number of patients with a
3.2.2 First revisions after primary hip surgery
A total of 37,444 first revisions of a hip prosthesis have procedures have been grouped by the year of the
been linked to a previous primary hip replacement primary operation. Figure 3.H4 (a) plots each Kaplan-
recorded in the registry between 2003 and 2020. Meier survival curve with a common origin, i.e. time zero
is equal to the year of operation. This illustrates that
Figures 3.H4 (a) and (b) illustrate temporal changes in revision rates increased between 2003 and 2007/8 and
the overall revision rates using Kaplan-Meier estimates; then declined between 2007/8 and 2020.
Figure 3.H4 (a) KM estimates of cumulative revision by year, in primary hip replacements.
7
Cumulative revision (%)
6
2003
2004
5 2005
2006
2007
4 2008
2009
2010
3 2011
2012
2013
2 2014
2015
2016
1
2017
2018
2019
0
2020
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
Years since primary
Figure 3.H4 (b) KM estimates of cumulative revision by year, in primary hip replacements plotted by year of primary.

2
0
2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021
Year of primary
Cumulative probability of revision after primary:

1 year 3 years 5 years 7 years 10 years 13 years 15 years
63
Figure 3.H4 (b) shows the same curves plotted against Table 3.H5 (page 65) provides Kaplan-Meier estimates
calendar time, where the origin of each curve is the of the cumulative percentage probability of first
year of operation. In addition, we have highlighted the revision, for any cause, firstly for all cases combined
revision rate at 1, 3, 5, 7, 10, 13 and 15 years. Figure and then by type of fixation and by bearing surface
3.H4 (b) separates each year, enabling changes in within each fixation group. The table shows updated
failure rates over time to be clearly identified. If revision estimates at 1, 3, 5, 10, 15 and 17 years from the
surgery and timing of revision surgery were static primary operation together with 95% Confidence
across time, it would be expected that all of the failure Intervals (95% CI). Estimates in blue italics indicate
curves would be the same shape and equally spaced; time points where fewer than 250 cases remained
departures from this indicate a change in the number at risk, meaning that the estimates are less reliable.
and timing of revision procedures. It is also very Kaplan-Meier estimates are not shown at all when the
clear that the 3, 5, 7, 10 and 13-year rate of revision numbers at risk fell below ten cases.
increases for operations occurring between 2003
and 2008 and then reduces for operations occurring Further revisions in the blue italicised groups would
between 2008 and 2020. The early increases may be be highly unlikely and, when they do occur, they may
partly a result of under-reporting in the earlier years of appear to have a disproportionate impact on the
the registry as this wasn’t mandatory at that time, but Kaplan-Meier estimate, i.e. the step upwards may
is also contributed to by the usage of metal-on-metal seem steeper. Furthermore, the upper 95% CI at
bearings, which peaked in 2008 and then fell (see these time points may be underestimated. Although
Table 3.H2 on page 51). a number of statistical methods have been proposed
to deal with this, they typically give different values
A similar pattern, although smaller in effect, is also and, as yet, there is no clear consensus for the large
observed in knees. Knees were not affected by the datasets presented here.
high revision rates of metal-on-metal bearings, and
thus the decreases observed since 2009 indicate a The revision rate of dual mobility bearings appears
broader improvement in outcomes overall. It appears higher up to five years across all fixation types than
that this secular decline in revision rates is still that of most of the unipolar bearing combinations,
ongoing. This improvement suggests the adoption of except metal-on-metal. The CoPoM dual mobility
evidence-based practice to which the NJR’s clinician bearings show lower revision estimates than the
feedback has contributed. For example, for a primary MoPoM combinations but with overlapping confidence
hip replacement performed in 2010, the 10-year intervals. The relatively small numbers at risk in the
revision rate is 3.7% (95% CI 3.5% - 3.8%) which dual mobility groups make it difficult to draw firm
is below the current NICE recommended threshold conclusions yet.
of 5% at ten years (NICE: Total hip replacement
and resurfacing arthroplasty for end-stage arthritis
of the hip. Technology appraisal guidance [TA304]
Published: 26 February 2014). Prior to 2014, the
revision threshold recommended by NICE was
10% at ten years (NICE: Guidance on the Selection
of Prostheses for Primary Total Hip Replacement.
Technology appraisal guidance [TA2] Published: 26
April 2000).
Table 3.H5 KM estimates of cumulative revision (95% CI) by fixation and bearing, in primary hip replacements.
Blue italics signify that fewer than 250 cases remained at risk at these time points.
Bearing Time since primary

Fixation surface N 1 year 3 years 5 years 10 years 15 years 17 years
All cases* 1,251,164 0.81 (0.79-0.83) 1.49 (1.47-1.51) 2.16 (2.13-2.19) 4.28 (4.23-4.33) 6.89 (6.78-6.99) 7.78 (7.62-7.95)
All cemented 391,414 0.57 (0.54-0.59) 1.09 (1.06-1.12) 1.52 (1.48-1.56) 2.90 (2.82-2.97) 5.15 (4.99-5.31) 6.12 (5.85-6.39)
Cemented and MoP 338,744 0.57 (0.55-0.60) 1.10 (1.07-1.14) 1.54 (1.50-1.59) 2.96 (2.88-3.03) 5.21 (5.05-5.38) 6.19 (5.91-6.47)
MoM 407 0.74 (0.24-2.29) 1.80 (0.86-3.73) 2.63 (1.42-4.84) 5.94 (3.86-9.10) 9.13 (6.04-13.69)
CoP 49,677 0.50 (0.44-0.57) 0.96 (0.87-1.05) 1.32 (1.21-1.44) 2.28 (2.10-2.48) 4.41 (3.93-4.95) 5.25 (4.52-6.10)
MoPoM 2,340 1.21 (0.83-1.77) 1.81 (1.32-2.49) 2.81 (2.06-3.84) 6.22 (2.56-14.73)
Others 246** 0.87 (0.22-3.43) 1.47 (0.47-4.57) 1.47 (0.47-4.57)
All uncemented 465,982 0.96 (0.93-0.99) 1.75 (1.71-1.79) 2.54 (2.49-2.59) 5.06 (4.97-5.15) 7.89 (7.70-8.09) 8.81 (8.48-9.16)
Uncemented and MoP 181,446 1.03 (0.98-1.07) 1.64 (1.58-1.71) 2.07 (2.00-2.14) 3.63 (3.51-3.76) 6.23 (5.91-6.58) 7.66 (7.00-8.39)
MoM 29,028 1.07 (0.95-1.19) 3.50 (3.29-3.72) 7.73 (7.43-8.05) 17.74 (17.28-18.20) 22.48 (21.87-23.09) 23.31 (22.54-24.11)
CoP 118,388 0.82 (0.77-0.87) 1.35 (1.28-1.42) 1.75 (1.66-1.83) 2.86 (2.70-3.02) 4.46 (4.12-4.83) 5.22 (4.62-5.90)
CoC 133,721 0.96 (0.91-1.01) 1.75 (1.68-1.83) 2.28 (2.20-2.36) 3.44 (3.32-3.56) 5.16 (4.86-5.48) 5.55 (5.14-5.99)
CoM 2,153 0.56 (0.32-0.98) 2.74 (2.13-3.53) 4.80 (3.96-5.80) 8.11 (6.99-9.39)
MoPoM 725 2.73 (1.75-4.26) 3.38 (2.22-5.13) 3.38 (2.22-5.13) 3.38 (2.22-5.13)
CoPoM 406 0.79 (0.26-2.43) 2.46 (0.90-6.61) 2.46 (0.90-6.61)
Others 115** 3.48 (1.32-9.00) 7.02 (3.57-13.55) 8.07 (4.27-14.95) 17.09 (10.33-27.54)
All hybrid 284,326 0.79 (0.76-0.83) 1.30 (1.26-1.34) 1.76 (1.71-1.82) 3.21 (3.11-3.31) 5.10 (4.88-5.34) 5.87 (5.49-6.26)
Hybrid and MoP 159,609 0.83 (0.79-0.88) 1.34 (1.28-1.40) 1.79 (1.72-1.86) 3.09 (2.97-3.22) 4.97 (4.68-5.27) 5.64 (5.19-6.14)
MoM 2,722 0.81 (0.53-1.23) 2.61 (2.06-3.30) 5.76 (4.90-6.76) 16.08 (14.59-17.69) 21.45 (19.50-23.55) 22.48 (20.13-25.06)
CoP 90,769 0.76 (0.70-0.82) 1.21 (1.13-1.29) 1.53 (1.44-1.63) 2.38 (2.19-2.59) 4.12 (3.48-4.88) 5.81 (4.52-7.46)
CoC 26,961 0.60 (0.52-0.70) 1.09 (0.97-1.23) 1.59 (1.45-1.76) 2.73 (2.51-2.97) 3.92 (3.56-4.32) 4.04 (3.62-4.51)
MoPoM 3,236 1.30 (0.95-1.77) 1.90 (1.42-2.54) 2.29 (1.67-3.13)
CoPoM 862 1.02 (0.51-2.03) 1.34 (0.66-2.69) 1.34 (0.66-2.69)
Others 166** 1.81 (0.59-5.51) 2.68 (1.00-7.09) 2.68 (1.00-7.09) 2.68 (1.00-7.09)
All reverse hybrid 32,596 0.85 (0.76-0.96) 1.47 (1.34-1.62) 1.96 (1.80-2.13) 3.44 (3.15-3.75) 6.39 (5.37-7.60) 6.68 (5.55-8.03)
Reverse hybrid and MoP 22,231 0.87 (0.76-1.01) 1.46 (1.30-1.63) 1.90 (1.72-2.11) 3.53 (3.16-3.94) 6.39 (5.21-7.83) 6.39 (5.21-7.83)
CoP 10,141 0.77 (0.62-0.96) 1.48 (1.25-1.74) 1.97 (1.70-2.29) 3.06 (2.63-3.56) 5.92 (4.24-8.23) 6.76 (4.63-9.84)
Others 224** 2.33 (0.97-5.50) 2.99 (1.34-6.61) 9.10 (5.06-16.08) 19.06 (11.74-30.09)
All resurfacing 40,081 1.20 (1.10-1.31) 2.96 (2.80-3.14) 5.20 (4.98-5.43) 10.41 (10.10-10.73) 13.96 (13.55-14.38) 14.79 (14.30-15.29)
Resurfacing and MoM 39,883 1.20 (1.10-1.31) 2.97 (2.80-3.14) 5.20 (4.99-5.43) 10.41 (10.10-10.73) 13.96 (13.56-14.38) 14.79 (14.31-15.29)
Others 198** 1.17 (0.29-4.60) 2.61 (0.98-6.85)
* Includes 36,765 with unsure fixation/bearing surface; **Wide CI because estimates are based on a small group size.
Note: Blank cells indicate the number at risk is below ten and thus estimates have been omitted as they are highly unreliable.
65
Figure 3.H5 KM estimates of cumulative revision in cemented primary hip replacements by bearing.
Blue italics in the numbers at risk table signify that fewer than 250 cases remained at risk at these time points.
20
15
10
0
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
Years since primary
Key: Numbers at risk
MoP 338,744 293,203 236,590 179,817 127,413 84,596 51,896 25,621 8,167
MoM 407 379 346 318 280 251 205 91 19
CoP 49,677 40,948 30,750 21,805 14,347 8,844 5,082 2,475 736
MoPoM 2,340 1,621 836 289 73 23 <4
Figures 3.H5 to 3.H8 (pages 66 to 69) illustrate the the revision rates for polyethylene-containing bearings
differences between the various bearing surface sub- do appear to differ beyond ten years, which may
groups for cemented, uncemented, hybrid and reverse represent the increased use of highly cross-linked
hybrid hips, respectively. Metal-on-metal bearings polyethylene over time. The long-term impacts of such
continue to perform worse than all other options changes will continue to be monitored. Dual mobility
regardless of fixation, apart from in cemented fixation bearings have higher early revision rates than other
where the results of the rarely used metal-on-metal options for cemented and uncemented fixation, this
combination are similar to metal-on-polyethylene- effect appears to persist in cemented fixation. Although
on-metal dual mobility. The failure rates for ceramic- a similar pattern is seen in hybrid fixation, the difference
on-polyethylene bearings remain consistently low or compared to alternatives is smaller. Given the relatively
equivalent to alternatives across all fixation options small numbers and the likely case mix selection, these
out to ten years and it is encouraging that these are patterns should continue to be monitored.
becoming more widely used with time. The trajectory of
Figure 3.H6 KM estimates of cumulative revision in uncemented primary hip replacements by bearing.
25
20

15
10
0
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
Years since primary
MoP 181,446 154,373 119,481 86,352 56,766 32,946 16,100 6,357 1,503
MoM 29,028 27,841 26,089 23,826 21,633 19,256 12,694 3,990 451
CoP 118,388 90,874 61,656 39,163 24,184 14,475 8,205 4,197 1,288
CoC 133,721 122,816 107,359 86,417 60,638 32,781 13,831 4,826 1,174
CoM 2,153 2,083 1,969 1,873 1,768 1,346 249 6 <4
MoPoM 725 396 245 135 75 27 <4 <4 <4
CoPoM 406 163 60 22 7 <4
Resurfacing 40,081 37,969 35,779 33,177 30,467 26,824 20,057 10,288 3,165
Figure 3.H7 KM estimates of cumulative revision in hybrid primary hip replacements by bearing.
25
20
15
10
0
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
Years since primary
MoP 159,609 127,678 93,373 64,110 42,377 26,243 14,722 6,481 1,827
MoM 2,722 2,527 2,211 1,977 1,758 1,534 1,175 524 171
CoP 90,769 62,384 36,021 18,163 8,500 4,778 2,513 1,207 398
CoC 26,961 25,146 22,340 18,629 14,189 9,706 5,856 2,761 588
MoPoM 3,236 1,704 717 210 50 5 <4 <4
CoPoM 862 331 111 37 4 <4
Figure 3.H8 illustrates the revision rate of metal-on- 13 years. After 13 years the numbers at risk are very
polyethylene and ceramic-on-polyethylene bearings low and therefore it is difficult to interpret survivorship at
used with reverse hybrid fixation in primary total hip greater than 13 years.
replacement. This shows little difference for the first
Figure 3.H8 KM estimates of cumulative revision in reverse hybrid primary hip replacements by bearing.
10

0
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
Years since primary
MoP 22,231 18,815 13,919 9,498 6,055 3,279 1,381 390 91
CoP 10,141 8,795 6,964 4,968 3,202 1,795 839 199 57
In Figures 3.H9 (a) and 3.H9 (b), the whole cohort has the age groups was greater in women than in men.
been sub-divided by age at primary operation and by Thus, for example, women under 55 years had higher
gender. Across the whole group, there was an inverse revision rates than their male counterparts in the same
relationship between the probability of revision and age band, whereas women aged 80 years and older
the age of the patient. A closer look at both genders had a lower revision rate than their male counterparts.
(Figure 3.H9 (a)) shows that the variation between
Figure 3.H9 (a) KM estimates of cumulative revision in all primary hip replacements by gender and age.
Males Females
15 15
12 12
9 9
6 6
3 3
0 0
0 2 4 6 8 10 12 14 16 0 2 4 6 8 10 12 14 16
Years since primary Years since primary
Key: Numbers at risk Numbers at risk
<55 74,837 64,235 52,225 40,967 30,714 21,757 13,504 6,517 2,120 75,417 64,482 51,909 40,128 29,518 20,148 12,344 5,954 1,871
55 to 59y 51,507 43,931 35,452 27,709 20,762 14,725 9,501 4,804 1,458 59,444 50,685 41,185 31,980 24,032 17,059 10,938 5,540 1,680
60 to 64y 71,766 61,470 50,139 39,423 29,879 20,661 12,405 5,719 1,671 90,132 77,782 63,393 50,052 38,050 26,300 15,938 7,304 2,201
65 to 69y 85,143 73,268 58,904 44,576 31,641 21,006 12,487 5,776 1,700 124,447 107,746 87,257 66,568 47,722 32,138 19,584 9,419 2,759
70 to 74y 87,603 72,910 56,334 42,062 29,955 19,487 10,627 4,385 1,054 142,998 120,318 93,750 70,777 50,918 33,546 19,132 8,365 2,242
75 to 79y 71,118 58,498 44,302 31,243 19,894 11,261 5,348 1,753 333 127,059 106,801 83,096 60,414 40,817 25,285 13,217 5,046 1,133
≥80 60,265 45,955 30,968 18,403 9,588 4,244 1,559 407 76 129,428 104,445 75,927 50,253 29,797 15,171 6,379 1,955 347
In Figure 3.H9 (b), primary total hip replacements metal-on-metal bearings; an age trend is seen in both
with metal-on-metal (or uncertain) bearing surfaces genders but rates for women are lower than for men
and resurfacings have been excluded. The revision across the entire age spectrum. The age-mediated
rates for the younger women are noticeably reduced disparity in revision rates for women appears to be
compared to the data in Figure 3.H9 (a) which includes increasing with longer follow-up.
Figure 3.H9 (b) KM estimates of cumulative revision in all primary hip replacements by gender and age,
excluding MoM and resurfacing.
Males Females
10 10
8 8

6 6
4 4
2 2
0 0
0 2 4 6 8 10 12 14 16 0 2 4 6 8 10 12 14 16
<55 54,747 45,369 34,568 24,687 15,897 8,976 4,644 2,350 840 64,503 54,156 42,240 31,241 21,398 13,007 6,913 3,281 1,079
55 to 59y 41,041 33,988 26,104 19,060 12,840 7,786 4,520 2,349 784 52,676 44,253 35,125 26,385 18,921 12,518 7,430 3,844 1,236
60 to 64y 61,191 51,466 40,714 30,663 21,797 13,647 7,624 3,697 1,156 82,766 70,783 56,878 44,093 32,653 21,590 12,599 5,958 1,890
65 to 69y 77,630 66,211 52,343 38,618 26,333 16,540 9,571 4,610 1,439 117,619 101,372 81,355 61,252 43,040 28,246 16,966 8,312 2,538
70 to 74y 82,499 68,276 52,186 38,383 26,766 16,900 9,056 3,849 944 136,742 114,637 88,651 66,288 47,022 30,455 17,247 7,737 2,140
75 to 79y 67,820 55,617 41,797 29,159 18,223 10,069 4,709 1,587 304 121,793 102,118 79,024 56,958 38,037 23,226 12,103 4,722 1,079
≥80 57,582 43,869 29,409 17,297 8,860 3,846 1,393 372 66 123,507 99,717 72,274 47,582 27,978 14,106 5,929 1,863 340
Table 3.H6 (page 72) further expands Table 3.H5 ceramic-on-polyethylene bearings in younger patients
to show separate estimates for males and females are striking. Resurfacing arthroplasty continues to
within each of four age bands, <55, 55 to 64, 65 show high failure rates in all groups, especially women.
to 74 and ≥75 years. Estimates are shown at 1, 3, Even in males under 55 years of age, resurfacing has
5, 10, 15 and 17 years after the primary operation. twice the revision rate of some alternatives out to 15
These estimates refine results shown in earlier reports, years. Dual mobility age and gender sub-groups are
but now with larger numbers of cases and therefore too small at this stage to provide firm conclusions on
generally narrower confidence intervals. The relatively relative revision rates.
good results obtained with ceramic-on-ceramic and
72
Table 3.H6 KM estimates of cumulative revision (95% CI) of primary hip replacements by gender, age group, fixation and bearing.
Males Females
Fixation Age at
Time since primary Time since primary
group/ primary
bearing (years) N 1 year 3 years 5 years 10 years 15 years 17 years N 1 year 3 years 5 years 10 years 15 years 17 years
0.94 2.09 3.29 6.45 9.72 10.56 0.89 2.11 3.47 7.56 11.48 12.89
All cases <55 74,837 75,417
(0.87-1.01) (1.98-2.20) (3.15-3.43) (6.22-6.68) (9.32-10.14) (10.01-11.15) (0.83-0.96) (2.00-2.22) (3.32-3.61) (7.31-7.82) (11.04-11.94) (12.23-13.58)
All 0.75 1.77 2.40 4.37 8.73 11.95 0.67 1.46 2.17 4.60 8.20 9.93
<55 5,303 8,178
cemented (0.55-1.02) (1.43-2.18) (1.99-2.89) (3.68-5.20) (7.24-10.52) (9.31-15.27) (0.52-0.88) (1.21-1.76) (1.85-2.54) (4.00-5.27) (7.05-9.52) (8.29-11.88)
0.94 2.26 3.08 5.65 11.47 14.58 0.85 1.82 2.52 5.41 9.15 11.14
MoP <55 2,162 3,727
(0.61-1.46) (1.69-3.00) (2.39-3.95) (4.53-7.03) (9.31-14.08) (11.35-18.62) (0.60-1.21) (1.42-2.32) (2.03-3.11) (4.56-6.41) (7.72-10.84) (9.11-13.58)
MoM <55 7 12 0
0.60 1.42 1.89 3.13 4.88 7.34 0.47 1.10 1.84 3.66 7.11 8.12
CoP <55 3,079 4,349
(0.38-0.94) (1.04-1.94) (1.43-2.51) (2.40-4.08) (3.38-7.00) (4.31-12.35) (0.30-0.72) (0.82-1.48) (1.44-2.35) (2.90-4.61) (5.19-9.70) (5.63-11.64)
2.50 2.50 2.50 4.67 4.67 4.67
MoPoM <55 41 75
(0.36-16.45) (0.36-16.45) (0.36-16.45) (1.53-13.83) (1.53-13.83) (1.53-13.83)
Others <55 14 0 0 15 0
All 0.97 2.17 3.38 6.64 9.71 10.17 0.90 2.00 3.18 6.42 9.71 11.76
<55 40,368 42,543
uncemented (0.88-1.07) (2.03-2.32) (3.19-3.58) (6.31-7.00) (9.05-10.41) (9.39-11.02) (0.81-0.99) (1.87-2.14) (3.01-3.37) (6.10-6.75) (9.08-10.38) (10.36-13.34)
0.99 1.94 2.78 4.88 8.34 8.34 0.95 1.80 2.48 4.22 8.07 11.51
MoP <55 5,140 6,234
(0.75-1.31) (1.58-2.39) (2.31-3.33) (4.10-5.82) (6.64-10.45) (6.64-10.45) (0.74-1.23) (1.49-2.18) (2.09-2.95) (3.55-5.02) (6.42-10.13) (8.46-15.56)
0.76 3.61 7.74 17.58 21.57 21.57 1.85 5.82 12.77 26.66 32.50 33.25
MoM <55 3,304 2,386
(0.51-1.12) (3.03-4.31) (6.87-8.72) (16.30-18.95) (20.00 - 23.24) (20.00 - 23.24) (1.38-2.48) (4.95-6.84) (11.48-14.18) (24.91-28.51) (30.42-34.68) (30.96-35.67)
1.00 1.82 2.51 3.48 5.50 7.31 0.88 1.46 2.02 3.31 5.42 9.79
CoP <55 11,092 11,437
(0.82-1.20) (1.57-2.12) (2.18-2.89) (2.93-4.13) (4.13-7.29) (5.15-10.33) (0.72-1.07) (1.24-1.72) (1.74-2.35) (2.76-3.96) (4.20-6.97) (6.13-15.45)
0.98 2.09 2.91 4.39 6.45 6.76 0.79 1.77 2.46 4.16 5.55 5.91
CoC <55 20,504 22,093
(0.86-1.13) (1.90-2.30) (2.68-3.17) (4.06-4.75) (5.56-7.48) (5.71-7.99) (0.68-0.92) (1.60-1.96) (2.25-2.68) (3.84-4.50) (4.95-6.21) (5.03-6.94)
1.02 4.61 7.84 12.39 4.93 8.79 11.99

CoM <55 197 270 0
(0.25-4.00) (2.42-8.67) (4.80-12.67) (8.40-18.07) (2.90-8.35) (5.93-12.93) (8.58-16.61)
2.63 2.63 2.63 1.72 1.72 1.72
MoPoM <55 42 59
(0.37-17.25) (0.37-17.25) (0.37-17.25) (0.24-11.62) (0.24-11.62) (0.24-11.62)
2.17 8.70
CoPoM <55 74 0 0 0 47
(0.31-14.45) (1.89-35.15)
7.14 7.14 5.88 11.76 11.76
Others <55 15 0 17
(1.04-40.92) (1.04-40.92) (0.85-34.98) (3.08-39.40) (3.08-39.40)
0.93 1.61 2.22 4.80 9.04 10.27 0.72 1.32 1.91 3.93 6.72 7.60
All hybrid <55 11,787 15,124
(0.77-1.12) (1.39-1.87) (1.93-2.54) (4.23-5.45) (7.75-10.54) (8.59-12.26) (0.60-0.87) (1.14-1.53) (1.68-2.18) (3.49-4.42) (5.78-7.80) (6.38-9.03)
1.55 2.57 3.36 5.95 10.68 13.01 0.82 1.80 2.37 4.31 9.95 11.47
MoP <55 1,842 2,623
(1.07-2.23) (1.91-3.44) (2.56-4.40) (4.62-7.65) (7.83-14.47) (9.15-18.32) (0.54-1.26) (1.34-2.43) (1.81-3.10) (3.39-5.48) (7.47-13.21) (8.44-15.50)
2.28 4.32 16.58 26.71 1.80 3.18 7.97 21.55 25.78
MoM <55 312 0 223
(1.09-4.73) (2.53-7.33) (12.76-21.40) (21.24-33.26) (0.68-4.72) (1.53-6.56) (5.03-12.51) (16.50-27.88) (19.61-33.45)
Note: All cases includes unclassified hip types.

Note: Rows with no data or only zeros have been suppressed.
Males Females
Fixation Age at
group/ primary
0.97 1.49 1.91 3.27 6.58 8.49 0.73 1.22 1.52 3.09 3.88 5.34
CoP <55 6,203 7,444
(0.75-1.25) (1.20-1.86) (1.53-2.38) (2.48-4.29) (3.87-11.09) (4.66-15.19) (0.56-0.96) (0.98-1.52) (1.22-1.88) (2.38-4.01) (2.72-5.52) (2.96-9.55)
0.59 1.17 1.75 3.26 5.48 5.48 0.56 1.06 1.66 3.01 4.81 5.29
CoC <55 3,241 4,645
(0.38-0.92) (0.85-1.62) (1.34-2.28) (2.61-4.07) (4.22-7.10) (4.22-7.10) (0.38-0.83) (0.80-1.41) (1.32-2.09) (2.48-3.65) (3.86-6.00) (4.05-6.90)
2.16 3.91 3.91 1.86 1.86 1.86
MoPoM <55 93 120
(0.55-8.37) (1.23-12.02) (1.23-12.02) (0.47-7.24) (0.47-7.24) (0.47-7.24)
CoPoM <55 80 0 0 0 57 0
8.33
Others <55 16 0 12
(1.22-46.10)
All reverse 1.21 2.33 2.83 5.83 13.47 1.18 1.88 3.02 5.12 7.57
<55 914 1,299
hybrid (0.67-2.18) (1.51-3.60) (1.88-4.25) (3.87-8.73) (8.25-21.58) (0.71-1.94) (1.25-2.82) (2.14-4.24) (3.73-7.03) (4.87-11.66)
0.58 3.84 3.84 8.67 20.55 0.38 0.84 1.95 3.62 9.11
MoP <55 179 275
(0.08-4.08) (1.74-8.36) (1.74-8.36) (4.25-17.28) (10.26-38.68) (0.05-2.70) (0.21-3.35) (0.73-5.17) (1.57-8.23) (3.71-21.43)
1.39 2.01 2.64 5.07 11.54 1.33 2.01 2.74 3.98 4.77
CoP <55 722 988
(0.75-2.58) (1.20-3.38) (1.64-4.24) (3.08-8.27) (5.75-22.41) (0.78-2.29) (1.29-3.14) (1.83-4.08) (2.62-6.04) (2.97-7.63)
2.78 5.82 18.31 40.96
Others <55 13 0 36
(0.40-18.13) (1.48-21.35) (7.91-39.11) (24.15-63.39)
All 0.83 2.17 3.85 7.38 10.21 10.53 1.25 4.99 9.23 19.67 24.75 25.85
<55 14,148 5,610
resurfacing (0.70-1.00) (1.94-2.43) (3.54-4.19) (6.93-7.86) (9.58-10.87) (9.84-11.26) (0.99-1.58) (4.45-5.60) (8.50-10.02) (18.64-20.75) (23.54-26.01) (24.46-27.29)
0.84 2.17 3.85 7.38 10.21 10.52 1.24 5.00 9.24 19.68 24.76 25.85
MoM <55 14,087 5,567
(0.70-1.00) (1.94-2.43) (3.54-4.19) (6.93-7.86) (9.58-10.87) (9.84-11.25) (0.98-1.57) (4.46-5.61) (8.51-10.03) (18.64-20.76) (23.55-26.02) (24.47-27.30)
2.63 2.56 2.56
Others <55 61 0 43
(0.37-17.25) (0.37-16.84) (0.37-16.84)
0.90 1.78 2.63 5.29 8.41 9.46 0.72 1.53 2.41 5.20 8.44 9.50
All cases 55 to 64 123,273 149,576
(0.85-0.95) (1.71-1.86) (2.53-2.73) (5.12-5.45) (8.10-8.72) (9.02-9.92) (0.68-0.76) (1.46-1.59) (2.33-2.50) (5.05-5.35) (8.17-8.73) (9.10-9.92)
All 0.67 1.47 2.01 4.05 7.62 9.06 0.47 1.06 1.63 3.40 6.69 7.86
55 to 64 17,813 29,700
cemented (0.56-0.80) (1.29-1.66) (1.80-2.24) (3.69-4.44) (6.91-8.39) (8.07-10.17) (0.40-0.56) (0.95-1.19) (1.48-1.79) (3.14-3.68) (6.17-7.25) (7.15-8.64)
0.71 1.73 2.36 4.75 8.65 10.05 0.52 1.22 1.88 3.80 7.25 8.48
MoP 55 to 64 11,200 19,797
(0.57-0.89) (1.50-2.00) (2.08-2.67) (4.29-5.26) (7.82-9.57) (8.93-11.29) (0.43-0.63) (1.07-1.39) (1.69-2.09) (3.49-4.14) (6.66-7.88) (7.69-9.36)
1.92 1.92 1.92 8.56
MoM 55 to 64 26 0 0 0 53
(0.27-12.88) (0.27-12.88) (0.27-12.88) (3.29-21.28)
0.61 0.99 1.37 2.36 4.39 5.52 0.37 0.72 1.07 2.21 4.58 5.20
CoP 55 to 64 6,489 9,689
(0.45-0.84) (0.77-1.27) (1.09-1.72) (1.90-2.94) (3.29-5.84) (3.83-7.91) (0.26-0.51) (0.57-0.92) (0.86-1.32) (1.81-2.70) (3.55-5.91) (3.94-6.87)
1.49 1.49 0.81 0.81 2.10
MoPoM 55 to 64 90 0 141
(0.21-10.13) (0.21-10.13) (0.11-5.63) (0.11-5.63) (0.51-8.43)
Others 55 to 64 8 20 0
73
74
Males Females
Fixation Age at
group/ primary
All 0.89 1.82 2.69 5.72 8.98 10.02 0.78 1.64 2.56 5.45 8.55 9.36
55 to 64 64,239 72,851
uncemented (0.82-0.97) (1.71-1.93) (2.56-2.83) (5.48-5.98) (8.47-9.51) (9.22-10.87) (0.72-0.85) (1.55-1.74) (2.43-2.68) (5.24-5.68) (8.11-9.01) (8.70-10.07)
0.95 1.87 2.44 4.58 8.13 10.43 0.74 1.55 1.95 3.70 6.44 8.41
MoP 55 to 64 15,649 19,710
(0.81-1.12) (1.66-2.11) (2.19-2.71) (4.13-5.07) (7.08-9.32) (8.53-12.71) (0.62-0.87) (1.38-1.74) (1.75-2.17) (3.36-4.06) (5.70-7.28) (6.75-10.44)
0.85 3.06 6.65 16.56 21.01 22.05 0.93 3.87 9.44 22.54 28.25 28.83
MoM 55 to 64 5,177 4,847
(0.64-1.14) (2.62-3.57) (6.00-7.37) (15.55-17.63) (19.74-22.35) (20.19-24.06) (0.70-1.24) (3.36-4.45) (8.65-10.31) (21.37-23.77) (26.78-29.79) (27.17-30.56)
0.81 1.37 1.79 3.00 5.09 5.09 0.64 1.22 1.64 2.94 4.38 4.56
CoP 55 to 64 19,285 21,872
(0.69-0.95) (1.20-1.56) (1.58-2.02) (2.62-3.43) (4.19-6.17) (4.19-6.17) (0.55-0.76) (1.07-1.39) (1.46-1.84) (2.60-3.33) (3.69-5.21) (3.80-5.46)
0.93 1.78 2.32 3.62 5.61 6.14 0.90 1.56 2.03 3.04 4.89 5.26
CoC 55 to 64 23,648 25,815
(0.81-1.06) (1.62-1.96) (2.13-2.53) (3.34-3.92) (4.84-6.49) (5.13-7.34) (0.79-1.02) (1.41-1.72) (1.86-2.22) (2.80-3.29) (4.26-5.61) (4.48-6.18)
0.63 2.87 5.18 7.40 0.43 1.95 3.28 6.85
CoM 55 to 64 317 465
(0.16-2.50) (1.50-5.44) (3.21-8.32) (4.93-11.04) (0.11-1.71) (1.02-3.72) (1.99-5.38) (4.87-9.60)
2.88 2.88 2.88
MoPoM 55 to 64 66 0 0 0 72
(0.73-11.05) (0.73-11.05) (0.73-11.05)
2.00 2.00
CoPoM 55 to 64 80 0 0 53
(0.28-13.36) (0.28-13.36)
5.88 5.88 5.88 7.14
Others 55 to 64 17 17 0
(0.85-34.98) (0.85-34.98) (0.85-34.98) (1.04-40.92)
0.84 1.50 2.05 3.69 6.15 6.86 0.60 1.17 1.68 3.50 5.33 6.27
All hybrid 55 to 64 23,177 34,174
(0.73-0.97) (1.34-1.67) (1.86-2.27) (3.34-4.07) (5.44-6.95) (5.84-8.06) (0.52-0.69) (1.05-1.30) (1.53-1.84) (3.22-3.80) (4.84-5.87) (5.45-7.19)
1.01 1.77 2.39 3.94 6.41 7.18 0.77 1.30 1.86 3.72 6.06 7.23
MoP 55 to 64 6,866 11,413
(0.80-1.28) (1.47-2.13) (2.03-2.82) (3.38-4.60) (5.34-7.68) (5.76-8.92) (0.62-0.95) (1.10-1.53) (1.61-2.15) (3.29-4.20) (5.29-6.93) (6.02-8.66)
0.77 4.25 7.32 16.08 21.57 25.93 0.71 3.40 8.07 22.48 28.16 28.16
MoM 55 to 64 388 426
(0.25-2.38) (2.63-6.85) (5.08-10.50) (12.63-20.35) (17.15-26.93) (17.79-36.87) (0.23-2.18) (2.03-5.68) (5.77-11.22) (18.58-27.05) (23.70-33.27) (23.70-33.27)
0.79 1.31 1.60 2.77 5.98 5.98 0.53 1.04 1.33 2.23 4.05 6.05
CoP 55 to 64 11,345 15,740
(0.64-0.97) (1.10-1.56) (1.35-1.91) (2.17-3.52) (4.05-8.80) (4.05-8.80) (0.43-0.66) (0.88-1.23) (1.13-1.56) (1.81-2.74) (2.82-5.80) (3.61-10.04)
0.67 1.16 1.73 2.79 4.13 4.13 0.42 1.00 1.43 2.59 3.03 3.03
CoC 55 to 64 4,323 6,279
(0.47-0.97) (0.88-1.53) (1.37-2.18) (2.28-3.41) (3.30-5.16) (3.30-5.16) (0.28-0.61) (0.78-1.28) (1.16-1.77) (2.18-3.08) (2.54-3.62) (2.54-3.62)
2.51 2.51 2.51 1.48 2.64 2.64
MoPoM 55 to 64 134 208
(0.81-7.61) (0.81-7.61) (0.81-7.61) (0.48-4.54) (0.91-7.53) (0.91-7.53)
1.09 1.09 1.15 1.15 1.15
CoPoM 55 to 64 97 91
(0.15-7.47) (0.15-7.47) (0.16-7.88) (0.16-7.88) (0.16-7.88)
6.67 5.88 5.88 5.88
Others 55 to 64 24 0 17
(0.97-38.74) (0.85-34.98) (0.85-34.98) (0.85-34.98)

Males Females
Fixation Age at
group/ primary
All reverse 0.95 1.84 2.44 4.00 9.68 0.90 1.69 2.36 4.08 8.78 8.78
55 to 64 2,596 4,054
hybrid (0.64-1.41) (1.37-2.47) (1.87-3.19) (3.01-5.31) (5.79-15.95) (0.65-1.25) (1.32-2.15) (1.91-2.93) (3.30-5.03) (6.17-12.42) (6.17-12.42)
0.80 1.37 2.17 4.56 11.43 1.18 1.91 2.78 5.05 11.45
MoP 55 to 64 1,020 1,722
(0.40-1.60) (0.80-2.35) (1.36-3.47) (2.81-7.37) (6.67-19.21) (0.76-1.82) (1.35-2.71) (2.05-3.78) (3.76-6.77) (7.54-17.20)
1.04 2.15 2.63 3.67 10.31 0.71 1.54 2.03 3.18 4.61
CoP 55 to 64 1,566 2,306
(0.64-1.70) (1.52-3.05) (1.90-3.65) (2.62-5.12) (3.54-28.05) (0.43-1.15) (1.10-2.16) (1.49-2.75) (2.35-4.30) (3.15-6.71)
5.56 14.14
Others 55 to 64 10 26 0 0
(0.80-33.36) (3.59-47.09)
All 1.20 2.35 3.75 6.95 9.62 10.49 1.66 4.48 8.53 17.33 22.54 23.99
55 to 64 11,949 4,289
resurfacing (1.02-1.42) (2.09-2.64) (3.42-4.12) (6.48-7.45) (8.98-10.29) (9.69-11.35) (1.32-2.09) (3.90-5.14) (7.72-9.41) (16.21-18.52) (21.20-23.95) (22.39-25.69)
1.20 2.34 3.74 6.94 9.60 10.48 1.67 4.50 8.55 17.35 22.56 24.01
MoM 55 to 64 11,911 4,263
(1.02-1.41) (2.08-2.63) (3.41-4.10) (6.47-7.43) (8.97-10.28) (9.68-11.34) (1.32-2.10) (3.91-5.16) (7.74-9.43) (16.23-18.54) (21.22-23.97) (22.40-25.71)
2.94
Others 55 to 64 38 26 0 0
(0.42-19.10)
0.88 1.52 2.09 4.05 6.75 7.70 0.70 1.26 1.80 3.44 5.39 6.05
All cases 65 to 74 172,746 267,445
(0.84-0.93) (1.46-1.58) (2.02-2.17) (3.92-4.18) (6.48-7.03) (7.23-8.21) (0.67-0.73) (1.22-1.31) (1.75-1.86) (3.35-3.54) (5.20-5.58) (5.76-6.36)
All 0.67 1.26 1.76 3.59 6.40 7.72 0.48 1.02 1.47 2.80 4.81 5.47
65 to 74 50,718 95,088
cemented (0.60-0.74) (1.16-1.36) (1.64-1.89) (3.38-3.81) (5.98-6.84) (6.97-8.55) (0.43-0.52) (0.96-1.09) (1.38-1.55) (2.66-2.94) (4.54-5.08) (5.08-5.89)
0.69 1.30 1.82 3.74 6.60 8.00 0.47 1.03 1.49 2.88 4.87 5.54
MoP 65 to 74 43,851 83,075
(0.62-0.77) (1.19-1.41) (1.69-1.96) (3.52-3.98) (6.16-7.06) (7.21-8.87) (0.42-0.52) (0.96-1.10) (1.40-1.58) (2.74-3.03) (4.60-5.16) (5.14-5.98)
1.72 3.54 3.54 6.02 0.99 0.99 3.09 5.33 7.98
MoM 65 to 74 58 101
(0.24-11.62) (0.90-13.45) (0.90-13.45) (1.95-17.80) (0.14-6.82) (0.14-6.82) (1.01-9.27) (2.25-12.34) (3.86-16.11)
0.49 0.90 1.27 2.15 4.17 4.17 0.51 1.00 1.24 1.90 4.05 4.70
CoP 65 to 74 6,606 11,468
(0.35-0.70) (0.69-1.17) (1.00-1.61) (1.71-2.69) (3.03-5.72) (3.03-5.72) (0.39-0.66) (0.82-1.21) (1.03-1.48) (1.58-2.29) (3.17-5.16) (3.34-6.59)
1.73 3.29 3.29 0.76 0.76 1.50
MoPoM 65 to 74 180 409
(0.56-5.28) (1.36-7.84) (1.36-7.84) (0.25-2.35) (0.25-2.35) (0.49-4.56)
3.13
Others 65 to 74 23 0 35
(0.45-20.18)
All 0.95 1.65 2.28 4.44 7.63 8.14 0.89 1.54 2.25 4.40 6.43 7.02
65 to 74 70,471 89,920
uncemented (0.88-1.03) (1.56-1.75) (2.16-2.40) (4.23-4.66) (7.09-8.21) (7.45-8.89) (0.83-0.96) (1.46-1.63) (2.14-2.35) (4.22-4.59) (6.08-6.80) (6.51-7.56)
0.92 1.57 1.95 3.76 6.90 7.36 0.92 1.46 1.87 3.26 5.05 5.77
MoP 65 to 74 31,387 44,045
(0.82-1.03) (1.44-1.72) (1.80-2.13) (3.47-4.09) (6.07-7.83) (6.35-8.53) (0.84-1.02) (1.35-1.58) (1.74-2.02) (3.04-3.50) (4.56-5.58) (5.01-6.64)
1.08 2.97 6.10 13.58 18.61 1.12 3.57 8.69 19.17 23.08
MoM 65 to 74 4,569 4,646
(0.82-1.42) (2.51-3.51) (5.43-6.85) (12.57-14.67) (17.01-20.34) (0.86-1.47) (3.07-4.15) (7.90-9.55) (18.02-20.38) (21.65-24.59)
0.82 1.25 1.50 2.28 4.25 4.56 0.74 1.21 1.54 2.63 3.90 3.90
CoP 65 to 74 18,656 22,446
(0.70-0.96) (1.10-1.43) (1.32-1.71) (1.98-2.64) (3.34-5.40) (3.51-5.90) (0.64-0.86) (1.07-1.37) (1.37-1.73) (2.32-2.98) (3.32-4.59) (3.32-4.59)
1.13 1.79 2.25 3.02 4.77 5.57 0.92 1.52 1.80 2.42 3.47 3.47
CoC 65 to 74 15,428 18,196
(0.98-1.31) (1.59-2.01) (2.02-2.50) (2.72-3.36) (3.95-5.77) (4.03-7.67) (0.79-1.07) (1.35-1.71) (1.61-2.01) (2.17-2.69) (2.96-4.06) (2.96-4.06)
75
76
Males Females
Fixation Age at
group/ primary
1.32 3.71 5.14 8.56 0.53 1.60 3.51 7.69
CoM 65 to 74 304 379
(0.50-3.49) (2.07-6.61) (3.13-8.38) (5.82-12.52) (0.13-2.11) (0.72-3.53) (2.05-5.96) (5.34-11.03)
4.45 6.44 6.44
MoPoM 65 to 74 72 0 0 0 139
(2.02-9.63) (3.24-12.58) (3.24-12.58)
CoPoM 65 to 74 46 0 45 0 0
8.33 12.50 12.50
Others 65 to 74 9 24
(2.15-29.39) (4.21-33.92) (4.21-33.92)
0.90 1.46 1.92 3.57 5.46 6.17 0.75 1.19 1.63 2.81 4.16 4.65
All hybrid 65 to 74 39,092 66,330
(0.81-1.00) (1.34-1.59) (1.77-2.08) (3.29-3.88) (4.89-6.09) (5.12-7.43) (0.69-0.82) (1.11-1.28) (1.53-1.75) (2.63-3.02) (3.79-4.56) (4.06-5.33)
0.89 1.47 1.95 3.64 5.74 6.21 0.77 1.24 1.69 2.88 4.14 4.22
MoP 65 to 74 22,101 40,403
(0.78-1.03) (1.31-1.65) (1.76-2.17) (3.29-4.02) (5.04-6.54) (5.13-7.50) (0.69-0.86) (1.13-1.36) (1.56-1.84) (2.66-3.13) (3.73-4.59) (3.78-4.71)
1.20 2.16 4.26 14.48 17.44 0.97 2.01 5.92 13.13 17.76
MoM 65 to 74 335 411
(0.45-3.16) (1.04-4.48) (2.49-7.23) (10.76-19.33) (13.13-22.97) (0.37-2.57) (1.01-3.99) (3.94-8.87) (9.95-17.23) (13.41-23.33)
0.88 1.34 1.58 2.45 3.11 5.66 0.70 1.10 1.37 1.89 3.46 6.01
CoP 65 to 74 13,369 20,878
(0.74-1.06) (1.15-1.57) (1.35-1.85) (1.94-3.08) (1.98-4.87) (2.27-13.74) (0.59-0.82) (0.96-1.26) (1.19-1.57) (1.61-2.20) (2.35-5.07) (3.21-11.11)
0.74 1.42 1.98 2.84 4.02 0.71 0.93 1.33 2.25 3.03
CoC 65 to 74 2,844 3,848
(0.49-1.14) (1.04-1.94) (1.51-2.59) (2.21-3.64) (3.07-5.27) (0.49-1.03) (0.67-1.29) (1.00-1.76) (1.76-2.88) (2.27-4.04)
3.16 4.48 4.48 1.76 2.45 3.34
MoPoM 65 to 74 313 620
(1.66-6.00) (2.39-8.29) (2.39-8.29) (0.95-3.24) (1.36-4.39) (1.70-6.52)
2.20 4.48 0.74 0.74 0.74
CoPoM 65 to 74 108 136
(0.55-8.52) (1.35-14.26) (0.10-5.10) (0.10-5.10) (0.10-5.10)
Others 65 to 74 22 0 33 0 0 0
All reverse 1.00 1.71 2.08 3.48 6.17 0.53 0.96 1.44 2.78 3.57
65 to 74 4,810 8,184
hybrid (0.75-1.32) (1.37-2.14) (1.68-2.56) (2.78-4.34) (4.20-9.00) (0.40-0.72) (0.77-1.21) (1.18-1.75) (2.28-3.38) (2.78-4.58)
1.23 1.94 2.43 4.08 7.43 0.55 0.98 1.39 2.91 3.25
MoP 65 to 74 3,389 6,083
(0.91-1.67) (1.51-2.49) (1.92-3.07) (3.17-5.24) (4.84-11.32) (0.39-0.77) (0.75-1.27) (1.11-1.76) (2.32-3.65) (2.53-4.16)
0.43 1.17 1.27 2.09 3.17 0.39 0.82 1.43 2.29 4.23
CoP 65 to 74 1,408 2,072
(0.19-0.96) (0.71-1.94) (0.78-2.06) (1.32-3.30) (1.53-6.51) (0.20-0.78) (0.50-1.34) (0.97-2.13) (1.52-3.46) (2.41-7.37)
6.90 6.90 12.07
Others 65 to 74 13 0 29
(1.77-24.86) (1.77-24.86) (3.94-33.73)
All 1.98 2.98 4.31 7.29 9.05 9.38 1.48 2.98 5.76 13.86 19.69
65 to 74 3,097 744
resurfacing (1.54-2.54) (2.44-3.65) (3.64-5.11) (6.37-8.33) (7.93-10.32) (8.12-10.83) (0.82-2.66) (1.97-4.49) (4.29-7.72) (11.50-16.65) (16.47-23.44)
1.99 2.99 4.32 7.30 9.06 9.39 1.52 3.04 5.84 13.95 19.80
MoM 65 to 74 3,088 727
(1.55-2.54) (2.44-3.66) (3.65-5.12) (6.38-8.34) (7.94-10.33) (8.13-10.84) (0.84-2.72) (2.01-4.58) (4.35-7.82) (11.58-16.76) (16.57-23.57)
Others 65 to 74 9 17 0

Males Females
Fixation Age at
group/ primary
1.00 1.56 2.02 3.37 5.01 5.23 0.72 1.13 1.48 2.46 3.52 3.82
All cases ≥75 131,383 256,487
(0.95-1.06) (1.49-1.63) (1.94-2.11) (3.22-3.53) (4.63-5.43) (4.75-5.76) (0.69-0.76) (1.09-1.17) (1.43-1.53) (2.37-2.55) (3.32-3.73) (3.43-4.25)
All 0.84 1.36 1.81 3.08 4.67 5.00 0.46 0.85 1.15 1.97 2.88 2.88
≥75 57,121 127,493
cemented (0.77-0.92) (1.26-1.46) (1.69-1.94) (2.87-3.31) (4.16-5.23) (4.34-5.76) (0.42-0.50) (0.80-0.90) (1.09-1.22) (1.87-2.08) (2.65-3.13) (2.65-3.13)
0.83 1.35 1.79 3.08 4.72 5.06 0.46 0.85 1.15 1.98 2.89 2.89
MoP ≥75 53,965 120,967
(0.76-0.92) (1.25-1.45) (1.67-1.92) (2.87-3.31) (4.20-5.31) (4.39-5.84) (0.42-0.50) (0.79-0.90) (1.08-1.22) (1.87-2.09) (2.65-3.14) (2.65-3.14)
2.56 11.07 3.22 3.22 6.16
MoM ≥75 46 0 104 0
(0.37-16.84) (3.59-31.36) (1.05-9.65) (1.05-9.65) (2.58-14.37)
0.82 1.41 1.79 2.37 2.37 0.40 0.70 0.92 1.39 2.21 2.21
CoP ≥75 2,670 5,327
(0.53-1.25) (1.00-1.98) (1.29-2.48) (1.64-3.44) (1.64-3.44) (0.26-0.62) (0.49-0.98) (0.66-1.27) (1.01-1.92) (1.38-3.52) (1.38-3.52)
1.82 2.76 5.04 0.94 1.56 2.60
MoPoM ≥75 412 992
(0.87-3.78) (1.49-5.08) (2.69-9.35) (0.49-1.79) (0.92-2.62) (1.57-4.28)
0.99 0.99
Others ≥75 28 0 103
(0.14-6.82) (0.14-6.82)
All 1.31 1.88 2.33 3.85 5.39 5.39 1.23 1.69 2.12 3.64 5.64 8.34
≥75 34,418 51,172
uncemented (1.19-1.43) (1.74-2.04) (2.16-2.51) (3.55-4.17) (4.71-6.17) (4.71-6.17) (1.14-1.33) (1.58-1.81) (1.99-2.26) (3.41-3.88) (4.94-6.43) (5.26-13.09)
1.37 1.95 2.31 3.36 5.35 1.23 1.62 1.99 3.28 5.34 8.05
MoP ≥75 23,118 36,163
(1.23-1.53) (1.77-2.14) (2.11-2.53) (3.04-3.72) (4.33-6.62) (1.12-1.35) (1.49-1.76) (1.84-2.15) (3.02-3.56) (4.37-6.53) (4.12-15.41)
1.02 1.91 3.74 8.55 10.14 1.34 3.05 4.94 9.54 12.16
MoM ≥75 1,700 2,399
(0.63-1.63) (1.34-2.70) (2.90-4.83) (7.11-10.28) (8.33-12.31) (0.95-1.89) (2.42-3.83) (4.12-5.92) (8.31-10.96) (10.54-14.02)
1.07 1.50 1.88 2.91 3.28 1.04 1.44 1.75 2.77 3.37 7.97
CoP ≥75 5,816 7,784
(0.83-1.37) (1.21-1.87) (1.52-2.32) (2.23-3.79) (2.37-4.53) (0.84-1.30) (1.19-1.74) (1.45-2.09) (2.27-3.37) (2.60-4.37) (2.58-23.21)
1.31 1.93 2.11 3.51 4.06 1.50 1.88 2.07 3.03 5.59
CoC ≥75 3,543 4,494
(0.99-1.75) (1.52-2.45) (1.67-2.65) (2.79-4.41) (3.09-5.32) (1.18-1.90) (1.52-2.33) (1.69-2.54) (2.44-3.76) (3.72-8.37)
2.76 4.26 0.77 0.77 1.88
CoM ≥75 88 0 0 133 0
(0.70-10.59) (1.39-12.63) (0.11-5.33) (0.11-5.33) (0.46-7.50)
5.08 5.08 5.08 2.50 3.57 3.57 3.57
MoPoM ≥75 102 173
(2.14-11.77) (2.14-11.77) (2.14-11.77) (0.94-6.53) (1.46-8.61) (1.46-8.61) (1.46-8.61)
2.63
CoPoM ≥75 40 21 0
(0.37-17.25)
0.94 1.50 1.95 3.25 4.77 0.77 1.14 1.52 2.23 2.93 3.72
All hybrid ≥75 32,107 62,535
(0.84-1.05) (1.36-1.65) (1.78-2.13) (2.92-3.63) (3.97-5.74) (0.71-0.84) (1.05-1.23) (1.41-1.63) (2.05-2.43) (2.49-3.43) (2.40-5.74)
0.92 1.51 1.96 3.20 4.86 0.78 1.16 1.49 2.21 3.00 3.87
MoP ≥75 24,748 49,613
(0.81-1.05) (1.36-1.68) (1.77-2.18) (2.84-3.60) (3.96-5.96) (0.70-0.86) (1.06-1.26) (1.38-1.62) (2.02-2.43) (2.52-3.57) (2.44-6.12)
0.88 1.41 1.97 10.12 0.50 1.61 4.24 8.93
MoM ≥75 228 399
(0.22-3.48) (0.46-4.33) (0.74-5.20) (5.85-17.19) (0.13-2.00) (0.73-3.56) (2.46-7.25) (5.96-13.27)
77
78
Males Females
Fixation Age at
group/ primary
1.06 1.50 1.98 2.62 2.62 0.77 1.04 1.56 1.64 1.64
CoP ≥75 5,865 9,925
(0.82-1.37) (1.19-1.89) (1.59-2.47) (1.97-3.49) (1.97-3.49) (0.61-0.97) (0.84-1.28) (1.27-1.90) (1.33-2.01) (1.33-2.01)
1.20 1.38 1.62 2.61 0.51 0.80 1.17 1.82
CoC ≥75 596 1,185
(0.57-2.49) (0.69-2.75) (0.84-3.10) (1.33-5.09) (0.23-1.13) (0.41-1.53) (0.66-2.08) (1.05-3.15)
0.42 0.96 0.96 0.70 1.04 1.58
MoPoM ≥75 539 1,209
(0.11-1.67) (0.36-2.56) (0.36-2.56) (0.35-1.39) (0.54-2.00) (0.83-2.97)
1.85 1.85
CoPoM ≥75 112 0 0 181
(0.60-5.68) (0.60-5.68)
4.35 4.35
Others ≥75 19 0 23
(0.62-27.07) (0.62-27.07)
All reverse 1.12 1.91 2.42 3.85 0.81 1.24 1.52 2.67 3.40
≥75 3,732 7,007
hybrid (0.83-1.52) (1.50-2.43) (1.93-3.04) (2.93-5.05) (0.62-1.05) (1.00-1.54) (1.24-1.87) (2.14-3.33) (2.63-4.40)
1.17 1.97 2.45 4.04 0.80 1.25 1.49 2.68 2.97
MoP ≥75 3,338 6,225
(0.85-1.60) (1.53-2.53) (1.93-3.12) (3.01-5.40) (0.60-1.05) (0.99-1.57) (1.20-1.85) (2.11-3.42) (2.30-3.83)
0.85 1.53 2.30 2.83 0.70 1.02 1.40 2.28

CoP ≥75 361 718
(0.28-2.62) (0.64-3.66) (1.10-4.80) (1.40-5.67) (0.29-1.68) (0.49-2.13) (0.73-2.68) (1.24-4.16)
3.36 3.36
Others ≥75 33 0 0 64
(0.85-12.80) (0.85-12.80)
All 1.91 2.43 4.19 6.97 6.97 3.13 6.58 6.58 12.42
≥75 212 32
resurfacing (0.72-5.00) (1.02-5.74) (2.11-8.23) (3.99-12.05) (3.99-12.05) (0.45-20.18) (1.68-23.91) (1.68-23.91) (3.95-35.37)
1.92 2.44 4.20 6.98 6.98 3.45 7.16 7.16 12.96
MoM ≥75 211 29
(0.72-5.03) (1.02-5.76) (2.11-8.25) (3.99-12.06) (3.99-12.06) (0.49-22.05) (1.84-25.75) (1.84-25.75) (4.19-36.23)

3.2.3 Revisions after primary hip e) Ceramic-on-polyethylene uncemented hip

constructs n=117,853
replacement: effect of head size for
selected bearing surfaces / fixation f) Ceramic-on-ceramic uncemented hip constructs
n=133,478
sub-groups
g) Metal-on-polyethylene hybrid hip constructs
This section looks at the effect of head size on n=159,417
the probability of revision following primary hip
h) Ceramic-on-polyethylene hybrid hip constructs
replacement. Fixation and bearing combinations with
n=90,583
greater than 10,000 uses are included, and head sizes
with less than 500 implantations within each group i) Ceramic-on-ceramic hybrid hip constructs
were excluded. n=26,447
This gave us 12 groups: j) Metal-on-polyethylene reverse hybrid hip

constructs n=21,541
a) Metal-on-polyethylene cemented hip constructs k) Ceramic-on-polyethylene reverse hybrid hip
n=338,706 constructs n=9,672
b) Ceramic-on-polyethylene cemented hip l) Metal-on-metal resurfacing n=39,271
constructs n=49,676
Figures 3.H10 (a) to 3.H10 (l) (on pages 80 to 91)
c) Metal-on-polyethylene uncemented hip show respective percentage cumulative probabilities
constructs n=180,942 of revision (Kaplan-Meier estimates) for various head
d) Metal-on-metal uncemented hip constructs sizes, for each of the groups with follow-up up to 17
n=28,528 years following the primary hip replacement.
Figure 3.H10 (a) KM estimates of cumulative revision of primary cemented MoP hip replacement by head
size (mm). Blue italics in the numbers at risk table signify that fewer than 250 cases remained at risk at these
time points.
10
8
0
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
Years since primary
22.25 34,815 32,659 29,819 26,301 22,186 17,270 12,271 7,498 3,116
26 18,689 17,736 16,470 14,739 12,411 9,655 6,877 4,045 1,487
28 198,010 176,816 146,941 114,103 81,021 52,947 30,985 13,484 3,427
30 730 692 629 494 293 190 108 42 <4
32 78,963 59,838 39,240 22,178 10,675 4,356 1,641 551 134
36 7,499 5,433 3,469 1,984 814 177 14 <4
In Figure 3.H10 (a), for cemented metal-on- rates over the entire duration of follow-up, but implants
polyethylene (MoP) hips, there was a statistically with head size 36mm had the worst failure rates in the
significant effect of head size (overall difference first nine years of follow-up. The numbers at risk for
P<0.001 by logrank test) on revision rates. Overall, patients who received 36mm heads after nine years
implants with head size 22.25mm had the worst failure are too small for meaningful comparison.
Figure 3.H10 (b) KM estimates of cumulative revision of primary cemented CoP hip replacement by
head size (mm). Blue italics in the numbers at risk table signify that fewer than 250 cases remained at risk
at these time points.
10

0
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
Years since primary
22.25 3,206 3,084 2,892 2,566 2,104 1,597 1,154 638 157
28 28,888 24,739 19,295 14,441 9,957 6,370 3,690 1,780 565
32 15,708 11,853 7,788 4,419 2,089 827 235 56 14
36 1,874 1,271 774 378 196 49 <4
Figure 3.H10 (b) shows revision rates for different head

sizes for cemented ceramic-on-polyethylene (CoP)
hips. There was a statistically significant effect of head
size (overall P<0.001) with 36mm heads having the
highest revision rates, followed by 22.25mm heads.
The lowest revision rates were achieved with 28mm
and 32mm heads.
Figure 3.H10 (c) KM estimates of cumulative revision of primary uncemented MoP hip replacement by
20
15
10
0
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
Years since primary
28 58,112 53,505 46,791 39,223 30,536 21,366 12,803 5,730 1,398
32 82,081 67,566 47,696 29,858 15,614 6,617 1,966 466 51
36 38,148 30,461 22,420 14,993 8,835 3,851 848 85 20
40 1,916 1,776 1,641 1,458 1,181 770 294 6 <4
44 685 637 569 513 384 213 84
Figure 3.H10 (c) shows revision rates for uncemented

metal-on-polyethylene (MoP) hips. Head sizes above
36mm had the highest revision rates.
Figure 3.H10 (d) KM estimates of cumulative revision of primary uncemented MoM hip replacement by
30
25

20
15
10
0
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
Years since primary
28 1,871 1,799 1,719 1,620 1,542 1,350 1,016 516 175
36 12,297 11,735 11,059 10,170 9,183 8,131 4,967 1,379 125
38 to 48 10,682 10,296 9,548 8,557 7,699 6,913 4,839 1,535 110
50 to 54 3,678 3,537 3,314 3,055 2,820 2,519 1,646 509 37
Figure 3.H10 (d) shows revision rates for uncemented

metal-on-metal (MoM) hips, with a statistically
significant difference between the head sizes overall
(P<0.001) with the lowest failure rates achieved with
the smallest head sizes.
Figure 3.H10 (e) KM estimates of cumulative revision of primary uncemented CoP hip replacement by
10
8
0
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
Years since primary
28 27,353 24,367 20,379 16,669 13,216 9,960 6,762 3,719 1,197
32 54,335 40,097 24,739 13,804 6,928 3,021 1,064 405 90
36 36,165 25,933 16,130 8,325 3,785 1,402 378 72
For uncemented ceramic-on-polyethylene (CoP) hips

(Figure 3.H10 (e)), there was a statistically significant
difference between the three head sizes shown
(P<0.001) with 28mm heads having higher revision
rates than 32mm and 36mm heads.
Figure 3.H10 (f) KM estimates of cumulative revision of primary uncemented CoC hip replacement by
10

0
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
Years since primary
28 13,445 12,799 11,965 10,957 9,489 7,357 5,142 2,770 859
32 36,485 33,246 28,610 22,827 16,054 8,682 3,906 1,331 285
36 77,544 70,976 61,496 48,127 31,867 15,461 4,769 725 30
40 5,361 4,938 4,487 3,796 2,680 1,017 14
44 643 627 594 520 400 203
Figure 3.H10 (f) shows revision rates for uncemented

ceramic-on-ceramic (CoC) hip replacements by head
size. There are statistically significant differences
between all five head sizes shown (P<0.001). The
larger the head size, the lower the revision rate of the
construct.
Figure 3.H10 (g) KM estimates of cumulative revision of primary hybrid MoP hip replacement by head
size (mm). Blue italics in the numbers at risk table signify that fewer than 250 cases remained at risk at
these time points.
10
8
0
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
Years since primary
22.25 862 706 581 462 342 259 203 149 69
26 886 845 749 667 545 411 296 178 62
28 46,852 42,277 36,096 29,322 23,193 17,175 11,363 5,589 1,636
32 70,542 52,754 34,011 19,381 9,582 4,371 1,602 360 41
36 38,674 29,454 20,443 12,973 7,641 3,405 1,017 201 19
40 1,601 1,468 1,335 1,170 975 561 220 4
Figure 3.H10 (g) shows revision rate for hybrid

MoP hip replacements by head size. There was a
statistically significant difference between the six head
sizes shown (P<0.001) with 22.25mm heads having
higher revision rates than the other heads. Beyond
ten years the numbers at risk are low so apparent
differences should be interpreted with caution.
Figure 3.H10 (h) KM estimates of cumulative revision of primary hybrid CoP hip replacement by head
these time points.
10

0
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
Years since primary
28 12,275 9,856 6,885 4,635 3,278 2,453 1,773 1,036 372
32 44,540 29,340 15,808 7,222 3,059 1,566 577 151 25
36 33,768 23,048 13,257 6,267 2,138 746 157 16
Figure 3.H10 (h) shows revision rates for hybrid

ceramic-on-polyethylene hip replacements by head
size. There were no statistically significant differences
in revision rates between 28mm, 32mm and 36mm
heads (P=0.06).
Figure 3.H10 (i) KM estimates of cumulative revision of primary hybrid CoC hip replacement by head
these time points.
4
0
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
Years since primary
28 5,367 5,148 4,777 4,247 3,623 2,793 1,879 1,066 349
32 13,341 12,664 11,494 9,886 7,630 5,312 3,134 1,294 184
36 7,739 6,843 5,652 4,156 2,750 1,556 843 401 55
Figure 3.H10 (i) shows revision rates for hybrid

ceramic-on-ceramic hip replacements by head size.
Bearings with 36mm heads had a higher revision rate
than 32mm and 28mm heads (P=0.009).
Figure 3.H10 (j) KM estimates of cumulative revision of primary reverse hybrid MoP hip replacement by
10

0
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
Years since primary
28 16,452 14,476 11,461 8,135 5,302 2,957 1,277 374 88
32 5,089 3,757 2,000 998 507 193 44 4
Figure 3.H10 (j) shows revision rates for reverse hybrid

metal-on-polyethylene hip replacements by head size.
There were no statistically significant differences in
revision rates between head sizes (P=0.09).
Figure 3.H10 (k) KM estimates of cumulative revision of primary reverse hybrid CoP hip replacement by
10
8
0
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
Years since primary
28 7,120 6,353 5,213 3,948 2,710 1,634 805 192 56
32 2,552 2,117 1,534 875 423 137 31 7 <4
Figure 3.H10 (k) shows revision rates for reverse

hybrid ceramic-on-polyethylene hip replacements
by head size. There were no statistically significant
differences in revision rates between head sizes
(P=0.24).
Figure 3.H10 (l) KM estimates of cumulative revision of primary resurfacing MoM hip replacement by
30
25

20
15
10
0
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
Years since primary
42 2,234 2,136 2,028 1,896 1,784 1,654 1,408 804 283
44 2,236 2,141 2,024 1,881 1,744 1,553 1,135 460 93
46 6,198 6,020 5,830 5,540 5,157 4,707 3,751 2,099 693
48 4,789 4,542 4,307 4,001 3,645 3,122 2,088 819 184
50 11,722 11,095 10,455 9,671 8,894 7,833 5,961 3,206 1,064
52 5,014 4,689 4,376 3,965 3,545 2,913 1,787 653 134
54 5,833 5,460 5,107 4,712 4,323 3,825 3,019 1,760 590
56 724 679 644 587 529 457 289 140 27
58 521 490 464 421 389 341 272 180 52
Figure 3.H10 (l) shows revision rates for resurfacing

metal-on-metal hip replacements by head size. There
is a strong trend to better implant survivorship with
larger head sizes.
3.2.4 Revisions after primary hip fallen below ten cases. No attempt has been made to
adjust for other factors that may influence the chance
surgery for the main stem / cup of revision, so the figures are unadjusted cumulative
brand combinations probabilities of revision. Given that the sub-groups
may differ in composition with respect to age and
As in previous reports, we only include only stem / cup
gender, the percentage of males and the median (IQR)
brand combinations with more than 2,500 procedures
of the ages are also shown in these tables.
for cemented, uncemented, hybrid and reverse hybrid
hips or more than 1,000 procedures in the case of Table 3.H7 shows Kaplan-Meier estimates of the
resurfacings. The figures in blue italics are at time cumulative percentage probability of revision of
points where fewer than 250 cases remained at risk; primary hip replacement (for any reason) for the main
no results are shown at all where the number had stem / cup brand constructs.
Table 3.H7 KM estimates of cumulative revision (95% CI) of primary hip replacement by fixation, and stem / cup
brand. Blue italics signify that fewer than 250 cases remained at risk at these time points.
Median
Time since primary
(IQR) age at Percentage
Stem:cup brand N primary (%) males 1 year 3 years 5 years 10 years 15 years 17 years
Cemented
C-Stem AMT
Cemented Stem[St] 0.63 1.25 1.55 2.75
3,380 75 (70 to 79) 31
: Charnley and Elite (0.41-0.96) (0.92-1.70) (1.17-2.05) (2.13-3.55)
Plus LPW[C]
C-Stem AMT
0.32 0.95 1.31 2.11
Cemented Stem[St] 4,794 77 (72 to 81) 33
(0.19-0.53) (0.69-1.30) (0.99-1.74) (1.60-2.79)
: Elite Plus Ogee[C]
C-Stem AMT
0.50 0.98 1.27 2.05
(0.39-0.63) (0.81-1.18) (1.06-1.53) (1.41-2.99)
: Marathon[C]
C-Stem Cemented
0.39 0.88 1.19 2.71 4.61 5.30
Stem[St] : Elite Plus 6,036 72 (66 to 77) 39
(0.26-0.58) (0.66-1.16) (0.93-1.52) (2.21-3.31) (3.70-5.74) (3.86-7.24)
Ogee[C]
C-Stem Cemented
0.44 0.89 1.28 2.13
Stem[St] : 9,649 68 (60 to 75) 41
(0.33-0.60) (0.71-1.11) (1.05-1.56) (1.73-2.63)
Marathon[C]
CPT CoCr Stem[St] 0.60 1.51 2.23 3.92 6.04
2,518 73 (67 to 79) 36
: Elite Plus Ogee[C] (0.36-1.00) (1.10-2.08) (1.71-2.91) (3.12-4.93) (4.51-8.08)
CPT CoCr Stem[St] 0.92 1.52 2.19 4.09 5.48 6.26
17,237 77 (71 to 81) 31
: ZCA[C] (0.79-1.08) (1.34-1.73) (1.96-2.45) (3.66-4.58) (4.71-6.37) (4.74-8.24)
Charnley Cemented
0.33 1.14 1.83 3.65 6.21 6.99
Stem[St] : Charnley 4,623 72 (66 to 78) 38
(0.20-0.54) (0.87-1.50) (1.47-2.28) (3.10-4.30) (5.32-7.23) (5.91-8.27)
Cemented Cup[C]
Charnley Cemented
0.38 1.21 1.87 3.71 6.16 6.88
Stem[St] : Charnley 10,495 73 (67 to 78) 38
(0.27-0.51) (1.02-1.45) (1.62-2.16) (3.32-4.14) (5.51-6.88) (6.07-7.79)
Ogee[C]
Charnley Cemented
Stem[St] : Charnley 0.38 0.76 1.16 2.47 3.95 4.93
6,979 74 (68 to 79) 29
and Elite Plus (0.26-0.55) (0.58-1.00) (0.93-1.46) (2.09-2.93) (3.34-4.66) (3.85-6.31)
LPW[C]
Exeter V40[St] :
0.64 1.39 2.05 2.75 4.72 5.16
Cenator Cemented 2,522 75 (69 to 80) 32
(0.39-1.04) (0.99-1.93) (1.55-2.70) (2.14-3.54) (3.66-6.08) (3.87-6.85)
Cup[C]
Exeter V40[St] :
0.65 1.25 1.49 2.19 3.44 4.75
Charnley and Elite 5,447 73 (68 to 79) 31
(0.47-0.91) (0.98-1.59) (1.19-1.87) (1.76-2.72) (2.44-4.85) (2.63-8.51)
Plus LPW[C]
Note: Blank cells indicate that the number at risk at the time shown has fallen below ten and thus estimates have been omitted as they are highly unreliable.
Note: [St] = Stem; [C] = Cup; [SL] = Shell liner.
Median
Time since primary
Exeter V40[St] :
0.33 0.65 0.87 1.50 3.30 3.59
Elite Plus Cemented 5,221 73 (67 to 79) 33
(0.20-0.53) (0.46-0.91) (0.65-1.17) (1.16-1.93) (2.40-4.52) (2.58-5.00)
Cup[C]
Exeter V40[St] : Elite 0.40 0.86 1.20 2.18 3.29 3.82
26,380 74 (69 to 80) 35
Plus Ogee[C] (0.33-0.48) (0.75-0.98) (1.07-1.35) (1.97-2.41) (2.90-3.73) (3.22-4.53)
Exeter V40[St]
: Exeter 0.57 1.02 1.39 2.40 4.13 5.42
94,901 74 (69 to 79) 34
Contemporary (0.52-0.62) (0.95-1.09) (1.31-1.47) (2.26-2.54) (3.79-4.51) (4.47-6.57)
Flanged[C]
Exeter V40[St]
: Exeter 0.95 1.62 2.14 3.95 7.43 8.69
29,111 75 (70 to 80) 32
Contemporary (0.84-1.06) (1.48-1.78) (1.97-2.32) (3.67-4.26) (6.74-8.18) (7.54-10.02)
Hooded[C]
Exeter V40[St] : 0.60 1.19 1.64 3.78 6.67 8.10
16,880 73 (67 to 79) 32
Exeter Duration[C] (0.49-0.73) (1.04-1.37) (1.45-1.85) (3.46-4.14) (6.06-7.33) (6.92-9.48)
Exeter V40[St] : 0.49 0.87 1.27 1.90
39,370 71 (63 to 78) 34
Exeter X3 Rimfit[C] (0.43-0.57) (0.78-0.98) (1.15-1.41) (1.55-2.33)
Exeter V40[St] : 0.47 0.86 1.11 1.68
8,599 71 (64 to 78) 35
Marathon[C] (0.35-0.65) (0.67-1.10) (0.88-1.41) (1.27-2.21)
Exeter V40[St] : 0.40 0.85 1.29 3.10 7.54 8.27
2,815 74 (68 to 80) 32
Opera[C] (0.22-0.71) (0.56-1.27) (0.92-1.80) (2.41-3.98) (5.61-10.10) (6.01-11.32)

MS-30[St] : Original
0.25 0.53 0.76 1.53 2.40 3.15
ME Muller Low 4,064 74 (68 to 80) 32
(0.13-0.46) (0.34-0.82) (0.52-1.11) (1.10-2.12) (1.55-3.72) (1.78-5.56)
Profile Cup[C]
Muller Straight
Stem[St] : Original 0.38 0.79 1.15 2.47 5.43 9.02
2,907 75 (70 to 80) 28
ME Muller Low (0.21-0.69) (0.52-1.20) (0.80-1.64) (1.85-3.29) (3.64-8.06) (4.94-16.17)
Profile Cup[C]
Stanmore Modular
Stem[St] : 0.45 1.09 1.51 2.42 4.45 5.66
5,436 75 (70 to 80) 29
Stanmore-Arcom (0.30-0.66) (0.84-1.40) (1.21-1.89) (1.99-2.94) (3.51-5.64) (4.14-7.71)
Cup[C]
Uncemented
Accolade[St] : 0.94 1.89 2.53 4.10 5.68 5.68
27,158 66 (59 to 73) 44
Trident[SL] (0.83-1.06) (1.73-2.06) (2.34-2.72) (3.83-4.39) (4.97-6.47) (4.97-6.47)
Accolade II[St] : 0.85 1.38 1.81
13,042 65 (57 to 72) 46
Trident[SL] (0.70-1.03) (1.17-1.64) (1.45-2.27)
Anthology[St] : R3 1.11 1.74 2.18 3.68
4,924 62 (53 to 69) 42
Cementless[SL] (0.85-1.45) (1.41-2.16) (1.78-2.66) (2.75-4.92)
Corail[St] : ASR 0.98 7.43 23.56 43.87 48.52
2,747 61 (54 to 67) 54
Resurfacing Cup[C] (0.68-1.43) (6.50-8.48) (22.00-25.22) (41.97-45.82) (46.42-50.66)
Corail[St] : Duraloc 0.75 1.68 2.47 5.45 11.02 12.89
4,001 70 (64 to 75) 39
Cementless Cup[SL] (0.53-1.08) (1.32-2.13) (2.02-3.01) (4.74-6.25) (9.74-12.46) (10.96-15.13)
Corail[St] : Pinnacle 0.93 1.55 2.21 2.85
10,711 66 (58 to 74) 41
Gription[SL] (0.76-1.13) (1.32-1.83) (1.87-2.61) (2.36-3.45)
Corail[St] : 0.77 1.47 2.14 4.55 7.39
170,277 66 (59 to 73) 45
Pinnacle[SL] (0.73-0.81) (1.41-1.53) (2.07-2.22) (4.41-4.70) (7.00-7.81)
Corail[St] : 0.59 1.09 1.64 2.94 3.71 4.65
3,281 67 (61 to 74) 40
Trilogy[SL] (0.37-0.92) (0.79-1.52) (1.24-2.16) (2.34-3.68) (2.79-4.91) (3.25-6.65)
Furlong Evolution
Cementless[St] : 1.33 1.83 2.13
5,133 62 (52 to 70) 39
Furlong HAC CSF (1.04-1.69) (1.48-2.25) (1.74-2.60)
Plus[SL]
Median
Time since primary
Furlong HAC 1.11 1.83 2.23 3.60 5.17 5.79
17,104 69 (63 to 76) 40
Stem[St] : CSF[SL] (0.96-1.28) (1.64-2.05) (2.01-2.46) (3.31-3.92) (4.73-5.64) (5.17-6.49)
Furlong HAC
1.10 1.75 2.03 2.68
Stem[St] : Furlong 24,545 66 (59 to 73) 45
(0.98-1.24) (1.59-1.92) (1.86-2.22) (2.45-2.93)
HAC CSF Plus[SL]
M/L Taper
1.23 1.78 2.15 2.64
Cementless[St] : 6,203 61 (53 to 68) 50
(0.99-1.54) (1.47-2.14) (1.80-2.56) (2.21-3.15)
Continuum[SL]
M/L Taper
1.26 2.05 2.32
Cementless[St] : 5,443 64 (55 to 71) 51
(1.00-1.60) (1.69-2.48) (1.93-2.80)
Trilogy IT[SL]
Metafix Stem[St] : 0.77 1.11 1.39 2.21
6,949 64 (56 to 70) 46
Trinity[SL] (0.59-1.01) (0.88-1.40) (1.10-1.74) (1.60-3.06)
Polarstem
0.75 1.00 1.20 1.77
Cementless[St] : R3 19,140 66 (58 to 72) 46
(0.63-0.88) (0.86-1.17) (1.03-1.40) (1.42-2.20)
Cementless[SL]
SL-Plus Cementless
1.46 3.13 4.49 7.34 9.07
Stem[St] : EP-Fit 3,797 66 (59 to 74) 43
(1.12-1.89) (2.62-3.74) (3.87-5.22) (6.47-8.32) (7.91-10.39)
Plus[SL]
Synergy Cementless
0.88 1.29 1.71 3.26
Stem[St] : R3 3,871 65 (57 to 71) 52
(0.63-1.24) (0.98-1.71) (1.33-2.20) (2.35-4.52)
Cementless[SL]
Taperloc Cementless
1.10 1.50 1.76 2.32
Stem[St] : Exceed 26,557 65 (58 to 72) 44
(0.98-1.23) (1.36-1.66) (1.61-1.94) (2.11-2.55)

ABT[SL]
Taperloc Complete
Cementless 0.86 1.34 1.60
3,737 63 (56 to 70) 49
Stem[St] : Exceed (0.61-1.21) (1.01-1.78) (1.23-2.10)
ABT[SL]
miniHip[St] : 1.37 2.12 2.42 3.55
2,501 56 (49 to 63) 45
Trinity[SL] (0.98-1.92) (1.61-2.78) (1.86-3.15) (2.17-5.78)
Hybrid
C-Stem AMT
0.67 1.16 1.56 2.90 3.03
(0.56-0.80) (1.00-1.34) (1.36-1.80) (2.37-3.55) (2.46-3.74)
: Pinnacle[SL]
CPCS[St] : R3 0.81 1.39 1.70
4,834 74 (68 to 79) 32
Cementless[SL] (0.59-1.11) (1.06-1.83) (1.28-2.27)
CPT CoCr Stem[St] 1.49 2.17 2.56 4.29
11,159 70 (62 to 77) 36
: Continuum[SL] (1.28-1.74) (1.90-2.48) (2.24-2.93) (3.21-5.73)
CPT CoCr Stem[St]
: Trabecular Metal 1.14 1.89 2.43 4.56 5.69
2,771 72 (64 to 79) 31
Modular Cementless (0.80-1.62) (1.43-2.49) (1.88-3.14) (3.50-5.94) (4.02-8.04)
Cup[SL]
CPT CoCr Stem[St] 1.24 1.81 2.28
11,497 69 (62 to 76) 37
: Trilogy IT[SL] (1.05-1.46) (1.57-2.09) (1.98-2.63)
CPT CoCr Stem[St] 0.90 1.45 2.18 3.86 5.28 5.28
24,410 71 (65 to 78) 36
: Trilogy[SL] (0.79-1.03) (1.30-1.61) (1.98-2.39) (3.51-4.26) (4.68-5.96) (4.68-5.96)
Exeter V40[St] :
0.27 0.74 1.20 2.25 3.98 3.98
ABG II Cementless 2,633 65 (59 to 73) 34
(0.13-0.56) (0.47-1.15) (0.84-1.71) (1.70-2.97) (3.08-5.13) (3.08-5.13)
Cup[SL]
Exeter V40[St] : 0.79 1.16 1.43 2.64 3.48
9,540 72 (65 to 78) 38
Pinnacle[SL] (0.63-0.99) (0.95-1.40) (1.19-1.71) (2.13-3.27) (2.65-4.57)
Exeter V40[St] : R3 0.73 1.23 1.53 2.09
3,092 72 (65 to 78) 31
Cementless[SL] (0.48-1.11) (0.88-1.72) (1.10-2.13) (1.27-3.43)
Median
Time since primary
Exeter V40[St] : 0.62 1.07 1.41 2.43 3.56 4.06
110,306 69 (61 to 76) 40
Trident[SL] (0.58-0.67) (1.01-1.14) (1.33-1.49) (2.28-2.59) (3.25-3.89) (3.34-4.93)
Exeter V40[St] : 0.57 0.89 1.24 2.20 3.38 4.18
14,776 70 (63 to 76) 40
Trilogy[SL] (0.46-0.70) (0.75-1.06) (1.07-1.44) (1.94-2.49) (2.94-3.90) (3.31-5.27)
Exeter V40[St] : 1.09 1.66 2.13 3.24
6,520 68 (60 to 75) 45
Tritanium[SL] (0.86-1.38) (1.35-2.03) (1.75-2.60) (2.56-4.08)
Taperfit Cemented 0.90 1.39 1.64
6,727 71 (65 to 77) 34
Stem[St] : Trinity[SL] (0.70-1.16) (1.12-1.73) (1.31-2.04)
Reverse hybrid

Corail[St] : Elite Plus 0.64 1.44 1.85 2.97 5.18
3,143 72 (65 to 77) 37
Ogee[C] (0.41-0.99) (1.07-1.93) (1.41-2.42) (2.31-3.82) (3.74-7.16)
Corail[St] : 0.63 1.08 1.33 2.12
16,069 70 (64 to 76) 39
Marathon[C] (0.52-0.77) (0.92-1.26) (1.14-1.54) (1.74-2.59)
Resurfacing
ASR Resurfacing 1.65 5.88 13.30 26.27 30.30
2,918 55 (49 to 60) 69
Cup (1.24-2.18) (5.08-6.80) (12.12-14.60) (24.70-27.93) (28.60-32.08)
Adept Resurfacing 1.09 2.41 4.40 7.90 11.93
3,806 54 (47 to 59) 75
Cup (0.80-1.48) (1.96-2.96) (3.77-5.13) (7.03-8.87) (10.11-14.06)
BHR Resurfacing 1.02 2.31 3.55 7.45 10.60 11.30
22,740 55 (48 to 60) 76
Cup (0.89-1.16) (2.12-2.51) (3.31-3.81) (7.09-7.83) (10.12-11.10) (10.74-11.88)
Conserve Plus 2.05 5.17 8.31 14.11 17.04 18.32
1,320 56 (50 to 61) 63
Resurfacing Cup (1.41-2.97) (4.10-6.51) (6.94-9.94) (12.32-16.13) (14.82-19.57) (15.18-22.02)
Cormet 2000 1.53 3.78 7.73 16.86 22.86 24.68
3,610 55 (48 to 60) 65
Resurfacing Cup (1.17-1.98) (3.20-4.45) (6.90-8.66) (15.67-18.14) (21.35-24.45) (22.89-26.58)
Durom Resurfacing 1.36 3.56 5.47 8.48 10.38
1,689 55 (49 to 60) 70
Cup (0.91-2.04) (2.78-4.56) (4.49-6.67) (7.24-9.92) (8.93-12.04)
1.95 3.37 5.58 10.21 13.73
Recap Magnum 1,693 54 (49 to 59) 73
(1.39-2.73) (2.61-4.35) (4.58-6.79) (8.84-11.78) (11.58-16.25)
Table 3.H8 further divides the data by stratifying there were more than 2,500 procedures for unipolar
for bearing surface. This table shows the estimated bearings, or more than 1,000 procedures for dual
cumulative percentage probability of revision for the mobility bearings.
resulting fixation / bearing sub-groups, provided
Table 3.H8 KM estimates of cumulative revision (95% CI) of primary hip replacement by fixation, stem / cup
brand, and bearing. Blue italics signify that fewer than 250 cases remained at risk at these time points.
Median
Time since primary
Bearing (IQR) age at Percentage
Stem:cup brand surface N primary (%) males 1 year 3 years 5 years 10 years 15 years 17 years
Cemented
C-Stem AMT
Cemented Stem[St] 0.63 1.26 1.56 2.78
MoP 3,352 75 (71 to 79) 31
: Charnley and Elite (0.41-0.97) (0.93-1.72) (1.18-2.07) (2.15-3.59)
Plus LPW[C]
C-Stem AMT
0.32 0.94 1.30 2.17
Cemented Stem[St] MoP 4,165 77 (73 to 82) 33
(0.18-0.55) (0.67-1.31) (0.96-1.76) (1.62-2.90)
: Elite Plus Ogee[C]
C-Stem AMT
0.46 0.99 1.34 1.83
(0.35-0.61) (0.81-1.22) (1.10-1.63) (1.36-2.45)
: Marathon[C]
C-Stem AMT
0.67 0.90 1.00 2.42
Cemented Stem[St] CoP 2,574 66 (60 to 71) 37
(0.42-1.08) (0.58-1.38) (0.64-1.53) (0.99-5.82)
: Marathon[C]
C-Stem Cemented
0.44 0.98 1.30 2.97 5.07 6.05
Stem[St] : Elite Plus MoP 5,077 73 (68 to 78) 38

(0.29-0.67) (0.73-1.31) (1.01-1.68) (2.41-3.67) (4.01-6.39) (4.17-8.73)
Ogee[C]
C-Stem Cemented
0.36 0.76 1.12 2.06
Stem[St] : MoP 5,467 73 (68 to 78) 37
(0.23-0.56) (0.55-1.04) (0.85-1.49) (1.53-2.78)
Marathon[C]
C-Stem Cemented
0.56 1.06 1.48 2.23
Stem[St] : CoP 4,182 59 (52 to 65) 46
(0.37-0.84) (0.78-1.45) (1.12-1.96) (1.67-2.99)
Marathon[C]
CPT CoCr Stem[St] 0.96 1.57 2.26 4.19 5.42 6.20
MoP 16,290 77 (72 to 82) 30
: ZCA[C] (0.82-1.12) (1.38-1.78) (2.01-2.53) (3.74-4.69) (4.65-6.31) (4.68-8.19)
Charnley Cemented
0.33 1.14 1.83 3.65 6.21 6.99
Stem[St] : Charnley MoP 4,623 72 (66 to 78) 38
(0.20-0.54) (0.87-1.50) (1.47-2.28) (3.10-4.30) (5.32-7.23) (5.91-8.27)
Cemented Cup[C]
Charnley Cemented
0.38 1.21 1.87 3.71 6.16 6.88
Stem[St] : Charnley MoP 10,495 73 (67 to 78) 38
(0.27-0.51) (1.02-1.45) (1.62-2.16) (3.32-4.14) (5.51-6.88) (6.07-7.79)
Ogee[C]
Charnley Cemented
Stem[St] : Charnley 0.38 0.76 1.16 2.47 3.95 4.93
MoP 6,979 74 (68 to 79) 29
and Elite Plus (0.26-0.55) (0.58-1.00) (0.93-1.46) (2.09-2.93) (3.34-4.66) (3.85-6.31)
LPW[C]
Exeter V40[St] :
0.68 1.24 1.49 2.39 3.91 5.24
Charnley and Elite MoP 4,297 75 (71 to 80) 28
(0.48-0.98) (0.94-1.64) (1.15-1.93) (1.88-3.05) (2.73-5.57) (2.98-9.13)
Plus LPW[C]
Exeter V40[St] :
0.35 0.63 0.82 1.42 2.80 3.12
Elite Plus Cemented MoP 4,926 74 (68 to 79) 32
(0.22-0.56) (0.44-0.89) (0.59-1.12) (1.08-1.86) (2.00-3.91) (2.18-4.47)
Cup[C]
Exeter V40[St] : Elite 0.38 0.86 1.20 2.18 3.31 3.88
MoP 23,869 75 (70 to 80) 34
Plus Ogee[C] (0.31-0.47) (0.74-0.99) (1.06-1.35) (1.97-2.42) (2.90-3.78) (3.24-4.65)
Exeter V40[St] : Elite 0.54 0.86 1.26 2.16 2.98 2.98
CoP 2,511 67 (61 to 72) 42
Plus Ogee[C] (0.31-0.93) (0.56-1.34) (0.86-1.84) (1.54-3.04) (2.06-4.29) (2.06-4.29)
*Inclusion criteria relaxed to show the newly identified dual mobility hips with at least 1,000 procedures.

Median
Time since primary
Exeter V40[St]
: Exeter 0.57 1.02 1.39 2.41 4.17 5.30
MoP 87,622 75 (70 to 80) 34
Contemporary (0.52-0.62) (0.95-1.09) (1.31-1.47) (2.27-2.56) (3.81-4.57) (4.34-6.48)
Flanged[C]
Exeter V40[St]
: Exeter 0.59 1.04 1.39 2.24 3.56
CoP 7,279 66 (61 to 72) 37
Contemporary (0.43-0.79) (0.83-1.32) (1.12-1.72) (1.81-2.78) (2.74-4.63)
Flanged[C]
Exeter V40[St]
: Exeter 0.96 1.62 2.14 3.93 7.39 8.57
MoP 27,198 76 (70 to 81) 32
Contemporary (0.85-1.08) (1.48-1.78) (1.97-2.33) (3.64-4.24) (6.68-8.17) (7.39-9.93)
Hooded[C]
Exeter V40[St] : 0.61 1.22 1.68 3.84 6.75 8.36
MoP 15,907 74 (68 to 79) 32
Exeter Duration[C] (0.50-0.75) (1.06-1.41) (1.49-1.90) (3.50-4.20) (6.12-7.44) (7.06-9.89)
Exeter V40[St] : 0.50 0.87 1.26 1.99
MoP 28,050 74 (68 to 79) 33
Exeter X3 Rimfit[C] (0.43-0.60) (0.76-1.00) (1.11-1.43) (1.52-2.59)
Exeter V40[St] : 0.46 0.88 1.29 1.71
CoP 11,320 63 (56 to 69) 38
Exeter X3 Rimfit[C] (0.35-0.61) (0.71-1.08) (1.07-1.56) (1.35-2.15)
Exeter V40[St] : 0.56 0.95 1.16 1.85
MoP 6,026 75 (70 to 80) 34
Marathon[C] (0.40-0.79) (0.72-1.26) (0.89-1.52) (1.32-2.57)
Exeter V40[St] : 0.28 0.63 1.01 1.30

CoP 2,573 62 (56 to 67) 40
Marathon[C] (0.13-0.58) (0.37-1.07) (0.63-1.61) (0.80-2.11)
Exeter V40[St] : 0.38 0.85 1.32 3.15 7.38 8.11
MoP 2,682 75 (69 to 80) 31
Opera[C] (0.20-0.70) (0.56-1.29) (0.94-1.85) (2.44-4.05) (5.48-9.91) (5.88-11.14)
MS-30[St] : Original
0.19 0.53 0.68 1.23 2.18 3.31
ME Muller Low CoP 2,609 71 (66 to 76) 32
(0.08-0.47) (0.31-0.92) (0.42-1.12) (0.80-1.91) (1.21-3.93) (1.53-7.07)
Profile Cup[C]
Stanmore Modular
Stem[St] : 0.41 1.09 1.56 2.50 4.16 5.17
MoP 4,966 75 (70 to 81) 30
Stanmore-Arcom (0.26-0.63) (0.83-1.42) (1.24-1.96) (2.04-3.06) (3.23-5.37) (3.65-7.29)
Cup[C]
Uncemented
Accolade[St] : 0.97 1.96 2.71 5.01 7.86
MoP 12,474 71 (64 to 76) 41
Trident[SL] (0.81-1.15) (1.73-2.22) (2.43-3.01) (4.56-5.51) (6.14-10.05)
Accolade[St] : 0.83 1.57 1.88 2.56 2.80
CoP 7,269 61 (55 to 67) 46
Trident[SL] (0.64-1.07) (1.30-1.89) (1.59-2.24) (2.12-3.09) (2.20-3.55)
Accolade[St] : 1.01 2.05 2.78 3.84 4.93 4.93
CoC 7,361 62 (55 to 68) 46
Trident[SL] (0.80-1.26) (1.75-2.40) (2.43-3.18) (3.41-4.33) (4.06-5.97) (4.06-5.97)
Accolade II[St] : 0.93 1.49 1.78
MoP 4,633 70 (64 to 76) 43
Trident[SL] (0.69-1.25) (1.14-1.96) (1.33-2.38)
Accolade II[St] : 0.85 1.38 1.76
CoP 7,740 62 (55 to 69) 48
Trident[SL] (0.66-1.09) (1.10-1.74) (1.24-2.50)
Anthology[St] : R3 1.17 1.82 2.11 2.45
MoP 3,912 63 (55 to 70) 39
Cementless[SL] (0.87-1.56) (1.44-2.30) (1.68-2.65) (1.95-3.09)
Corail[St] : ASR 0.98 7.43 23.56 43.87 48.52
MoM 2,747 61 (54 to 67) 54
Resurfacing Cup[C] (0.68-1.43) (6.50-8.48) (22.00-25.22) (41.97-45.82) (46.42-50.66)
Corail[St] : Duraloc
0.63 1.47 2.30 5.32 10.63 12.33
Cementless MoP 3,679 70 (65 to 75) 38
(0.42-0.94) (1.12-1.92) (1.85-2.85) (4.60-6.16) (9.27-12.17) (10.38-14.62)
Cup[SL]
Corail[St] : Pinnacle 1.08 1.63 2.22 3.16
MoP 3,909 74 (68 to 79) 37
Gription[SL] (0.80-1.47) (1.26-2.12) (1.70-2.88) (2.28-4.38)
Median
Time since primary
Corail[St] : Pinnacle 0.64 1.27 1.71
CoP 4,406 63 (57 to 70) 43
Gription[SL] (0.44-0.94) (0.94-1.71) (1.24-2.36)
Corail[St] : 0.79 1.27 1.56 2.77 4.55
MoP 68,083 71 (65 to 77) 41
Pinnacle[SL] (0.73-0.86) (1.19-1.36) (1.46-1.66) (2.59-2.96) (4.02-5.14)
Corail[St] : 0.88 2.45 5.20 13.33 17.76
MoM 11,884 67 (60 to 74) 47
Pinnacle[SL] (0.73-1.06) (2.18-2.74) (4.80-5.62) (12.69-13.99) (16.84-18.73)
Corail[St] : 0.66 1.06 1.43 2.42 3.51
CoP 44,886 64 (57 to 69) 47
Pinnacle[SL] (0.59-0.74) (0.97-1.17) (1.30-1.56) (2.14-2.75) (2.55-4.83)
Corail[St] : 0.83 1.78 2.42 3.77 5.84
CoC 43,589 59 (52 to 66) 49
Pinnacle[SL] (0.75-0.92) (1.65-1.91) (2.28-2.57) (3.56-3.99) (5.14-6.64)
Furlong Evolution
Cementless[St] : 1.24 1.66 2.01
CoC 4,422 60 (50 to 69) 39
Furlong HAC CSF (0.95-1.62) (1.31-2.10) (1.61-2.53)
Plus[SL]
Furlong HAC 1.37 2.19 2.54 4.24 5.67 7.48
MoP 8,079 73 (67 to 78) 39
Stem[St] : CSF[SL] (1.14-1.65) (1.89-2.53) (2.21-2.91) (3.77-4.76) (4.96-6.47) (5.68-9.81)
Furlong HAC 0.79 1.36 1.77 2.74 4.30 4.64
CoP 7,374 67 (61 to 73) 41
Stem[St] : CSF[SL] (0.61-1.02) (1.12-1.66) (1.49-2.10) (2.37-3.17) (3.72-4.97) (3.98-5.40)
Furlong HAC
1.66 2.32 2.82 3.98
Stem[St] : Furlong MoP 5,877 74 (69 to 79) 39
(1.37-2.03) (1.96-2.74) (2.41-3.30) (3.38-4.69)
HAC CSF Plus[SL]
Furlong HAC
0.91 1.58 1.83 2.72
Stem[St] : Furlong CoP 3,347 67 (62 to 72) 46
(0.63-1.29) (1.20-2.07) (1.41-2.37) (2.09-3.55)
HAC CSF Plus[SL]
Furlong HAC
0.93 1.57 1.78 2.23
Stem[St] : Furlong CoC 15,321 63 (56 to 69) 47
(0.79-1.10) (1.38-1.78) (1.58-2.01) (1.99-2.51)
HAC CSF Plus[SL]
Metafix Stem[St] : 0.68 0.91 1.13 1.13
CoP 3,077 64 (57 to 70) 47
Trinity[SL] (0.44-1.05) (0.61-1.36) (0.77-1.67) (0.77-1.67)
Metafix Stem[St] : 0.77 1.14 1.38 2.22
CoC 2,778 60 (52 to 66) 45
Trinity[SL] (0.50-1.18) (0.80-1.63) (0.98-1.94) (1.51-3.27)
Polarstem
0.78 1.04 1.26 2.11
Cementless[St] : R3 MoP 17,313 66 (59 to 73) 46
(0.66-0.92) (0.89-1.22) (1.07-1.49) (1.56-2.85)
Cementless[SL]
Synergy
Cementless 0.91 1.20 1.45 1.90
MoP 3,099 66 (58 to 72) 51
Stem[St] : R3 (0.63-1.31) (0.86-1.66) (1.07-1.96) (1.40-2.55)
Cementless[SL]
Taperloc
Cementless 1.29 1.80 2.04 2.73
MoP 8,493 72 (66 to 77) 40
Stem[St] : Exceed (1.07-1.56) (1.53-2.11) (1.75-2.38) (2.33-3.21)
ABT[SL]
Taperloc
Cementless 0.81 1.02 1.16 1.77
CoP 5,726 65 (58 to 70) 45
Stem[St] : Exceed (0.61-1.08) (0.79-1.33) (0.90-1.49) (1.33-2.34)
ABT[SL]
Taperloc
Cementless 1.09 1.52 1.84 2.30
CoC 12,325 61 (54 to 67) 47
Stem[St] : Exceed (0.92-1.29) (1.32-1.76) (1.61-2.10) (2.01-2.63)
ABT[SL]

Median
Time since primary
Hybrid
C-Stem AMT
0.70 1.27 1.71 2.73
(0.55-0.89) (1.04-1.54) (1.42-2.06) (2.04-3.64)
: Pinnacle[SL]
C-Stem AMT
0.65 0.97 1.06 1.44
Cemented Stem[St] CoP 7,022 67 (61 to 72) 42
(0.49-0.88) (0.75-1.25) (0.82-1.39) (0.93-2.21)
: Pinnacle[SL]
CPCS[St] : R3 0.79 1.40 1.58
MoP 4,447 74 (69 to 80) 31
Cementless[SL] (0.56-1.10) (1.05-1.87) (1.17-2.13)
CPT CoCr Stem[St] 1.61 2.22 2.67 4.66
MoP 5,721 75 (70 to 80) 34
: Continuum[SL] (1.31-1.97) (1.84-2.66) (2.20-3.22) (3.05-7.07)
CoP 3,947 65 (59 to 71) 39
: Continuum[SL] (1.06-1.81) (1.68-2.68) (1.85-2.98)
MoP 5,530 74 (69 to 79) 34
: Trilogy IT[SL] (1.20-1.85) (1.75-2.56) (2.23-3.26)
CoP 4,617 65 (59 to 71) 40
: Trilogy IT[SL] (0.77-1.37) (1.26-2.05) (1.63-2.69)

CPT CoCr Stem[St] 0.88 1.47 2.29 4.13 5.51 5.51
MoP 14,615 73 (67 to 79) 35
: Trilogy[SL] (0.74-1.04) (1.28-1.69) (2.04-2.56) (3.71-4.60) (4.86-6.26) (4.86-6.26)
CPT CoCr Stem[St] 0.95 1.43 2.00 2.57
CoP 9,272 69 (62 to 75) 37
: Trilogy[SL] (0.77-1.17) (1.20-1.71) (1.69-2.37) (2.14-3.09)
Exeter V40[St] : 0.83 1.21 1.50 2.48 3.33
MoP 6,265 75 (70 to 80) 31
Pinnacle[SL] (0.63-1.09) (0.96-1.53) (1.21-1.86) (1.95-3.14) (2.44-4.53)
Exeter V40[St] : 0.62 0.88 1.05 2.82
CoP 3,006 65 (59 to 71) 53
Pinnacle[SL] (0.39-0.98) (0.59-1.31) (0.71-1.54) (1.66-4.77)
Exeter V40[St] : 0.66 1.13 1.45 2.52 3.68
MoP 56,398 74 (68 to 79) 37
Trident[SL] (0.59-0.73) (1.04-1.23) (1.34-1.56) (2.31-2.76) (3.22-4.20)
Exeter V40[St] : 0.58 0.93 1.19 1.85 2.68
CoP 38,664 65 (58 to 71) 42
Trident[SL] (0.50-0.66) (0.83-1.04) (1.06-1.33) (1.58-2.18) (1.58-4.50)
Exeter V40[St] : 0.54 1.06 1.56 2.69 3.83 4.08
CoC 13,021 59 (53 to 65) 44
Trident[SL] (0.43-0.68) (0.90-1.25) (1.36-1.79) (2.40-3.01) (3.38-4.34) (3.45-4.84)
Exeter V40[St] : 1.14 1.88 2.11
MoPoM 1,442 75 (67 to 81) 33
Trident[SL]* (0.69-1.89) (1.18-2.99) (1.32-3.35)
Exeter V40[St] : 0.56 0.88 1.27 2.23 3.46 3.84
MoP 11,878 71 (65 to 77) 40
Trilogy[SL] (0.44-0.71) (0.72-1.07) (1.07-1.49) (1.94-2.56) (2.95-4.05) (3.15-4.68)
Exeter V40[St] : 0.55 0.93 1.14 2.00 3.07 4.92
CoP 2,758 63 (58 to 69) 43
Trilogy[SL] (0.33-0.90) (0.63-1.37) (0.80-1.63) (1.49-2.67) (2.22-4.25) (2.77-8.69)
Exeter V40[St] : 1.08 1.66 2.15 3.54
CoP 3,536 64 (57 to 70) 47
Tritanium[SL] (0.79-1.49) (1.26-2.19) (1.64-2.83) (2.32-5.37)
MoP 3,386 75 (70 to 80) 33
CoP 2,584 68 (62 to 74) 36
Reverse hybrid
Corail[St] : 0.64 1.07 1.31 2.12
MoP 11,242 73 (68 to 78) 38
Marathon[C] (0.51-0.81) (0.88-1.29) (1.10-1.57) (1.67-2.70)
Corail[St] : 0.59 1.10 1.35 2.14
CoP 4,827 62 (56 to 68) 41
Marathon[C] (0.41-0.86) (0.83-1.46) (1.04-1.75) (1.49-3.07)
3.2.5 Revisions for different causes time and so a small number of revisions associated
with a few primaries as late as 2011 still had revisions
after primary hip replacement reported on MDSv1 and MDSv2 of the data collection
Overall, 37,444 (3.0%) of the 1,251,164 primary hip forms. Restricting our analyses to primaries from 2008
replacements had an associated first revision. The onwards, as done in recent annual reports, ensures
most common indications for revision were aseptic that >99% of revisions were recorded on later forms
loosening (9,190), dislocation / subluxation (6,503), (MDSv3 onwards). It was noted that only 2,648 of
adverse soft tissue reaction to particulate debris the 5,698 instances (46.5%) of adverse reactions to
(5,698, a figure that is likely to be an underestimate particulate debris would thus be included, i.e. 3,050 of
due to changes in MDS collection, see later), the earlier cases are therefore missing. Therefore, two
periprosthetic fracture (5,696), infection (5,660) and sets of PTIRs are presented: one set for all primary hip
pain (4,916). Pain was not usually cited alone; in replacements, which are likely to be underestimates,
3,342 out of the 4,916 instances (68.0%), it was and the other set for all primary hip replacements
cited together with one or more other indications. performed since the beginning of 2008, which has
Associated PTIRs for these and the other indications better ascertainment but does not include the cases
are shown in Table 3.H9. Here, implant wear denotes with the longest follow-up.
wear of the polyethylene component, wear of the
Table 3.H9 reports revision by indication with further
acetabular component or dissociation of the liner.
breakdowns by hip fixation and bearing. Metal-
The number of adverse reactions to particulate debris on-metal (irrespective of the type of fixation) and
is likely to be underestimated because this was not resurfacings seem to have the highest PTIRs for
requested (i.e. it was not available as an indication for both aseptic loosening and pain but ceramic-on-
revision) on the revision data collection forms in the metal has similar rates. Metal-on-metal bearings
early phase of the registry, i.e. was not included in have the highest incidence of adverse reaction to
MDSv1 and MDSv2. Some of these cases may have particulate debris. Although the numbers are relatively
recorded the indication for revision as ‘other’ but this small in comparison to other groups, dual mobility
is not definitively known. Adoption of the later revision bearings appear to have high PTIRs for revision for
report forms (MDSv3 onwards) was staggered over periprosthetic fracture and infection.
Table 3.H9 PTIR estimates of indications for hip revision (95% CI) by fixation and bearing.
Adverse reaction to
particulate debris
for primaries from
Number of revisions per 1,000 prosthesis-years for: 1.1.2008***
Pros- Number of
thesis- Head/ Adverse Prosthesis- revisions
years Peripros- socket reaction to years per 1,000
Fixation/ at risk Aseptic Dislocation/ thetic Malalign- Implant Implant size particulate at risk prosthesis-
bearing type (x1,000) All causes loosening Pain Subluxation Infection fracture ment Lysis wear fracture mismatch debris** (x1,000) years
4.59 1.13 0.60 0.80 0.69 0.70 0.31 0.27 0.25 0.14 0.03 0.70 0.45
All cases* 8,159.6 5,913.6
(4.54-4.64) (1.10-1.15) (0.59-0.62) (0.78-0.82) (0.68-0.71) (0.68-0.72) (0.30-0.32) (0.26-0.28) (0.24-0.27) (0.14-0.15) (0.03-0.03) (0.68-0.72) (0.43-0.47)
All 3.22 1.05 0.25 0.80 0.66 0.53 0.18 0.21 0.18 0.08 0.01 0.03 0.03
2,660.6 1,730.7
cemented (3.15-3.29) (1.01-1.09) (0.23-0.27) (0.77-0.84) (0.63-0.70) (0.50-0.56) (0.16-0.20) (0.19-0.23) (0.17-0.20) (0.07-0.09) (0.01-0.02) (0.03-0.04) (0.02-0.03)
Cemented and
3.27 1.08 0.25 0.82 0.65 0.54 0.18 0.22 0.19 0.07 0.01 0.03 0.02
MoP 2,348.8 1,488.7
(3.19-3.34) (1.04-1.13) (0.23-0.27) (0.79-0.86) (0.62-0.69) (0.51-0.57) (0.17-0.20) (0.20-0.24) (0.17-0.21) (0.06-0.08) (0.01-0.02) (0.02-0.04) (0.02-0.03)
6.49 2.16 0.24 1.20 0.72 0.96 0.96 0.48 0.96 0.72
MoM 4.2 0 0 0.9 0
(4.45-9.46) (1.12-4.16) (0.03-1.71) (0.50-2.89) (0.23-2.23) (0.36-2.56) (0.36-2.56) (0.12-1.92) (0.36-2.56) (0.23-2.23)
2.70 0.78 0.20 0.62 0.70 0.38 0.15 0.14 0.11 0.10 0.04 0.04
CoP 299.0 0 232.5
(2.52-2.89) (0.69-0.89) (0.16-0.26) (0.54-0.72) (0.61-0.80) (0.32-0.46) (0.11-0.20) (0.10-0.19) (0.08-0.16) (0.07-0.15) (0.02-0.07) (0.02-0.07)
6.45 0.76 0.13 1.01 2.15 2.15 0.13 0.13
MoPoM 7.9 0 0 0 0 7.9 0
(4.90-8.49) (0.34-1.69) (0.02-0.90) (0.51-2.02) (1.34-3.46) (1.34-3.46) (0.02-0.90) (0.02-0.90)
All 5.37 1.31 0.73 0.80 0.69 0.68 0.42 0.28 0.34 0.19 0.05 1.13 0.73
3,047.5 2,443.9
uncemented (5.29-5.45) (1.27-1.35) (0.70-0.76) (0.77-0.83) (0.66-0.72) (0.66-0.71) (0.39-0.44) (0.26-0.30) (0.32-0.36) (0.17-0.20) (0.04-0.06) (1.09-1.17) (0.70-0.77)
Uncemented and
4.20 1.03 0.40 0.97 0.65 0.87 0.38 0.22 0.42 0.09 0.04 0.19 0.19
MoP 1,125.8 920.9
(4.08-4.32) (0.98-1.10) (0.36-0.44) (0.92-1.03) (0.60-0.70) (0.82-0.93) (0.35-0.42) (0.19-0.25) (0.39-0.46) (0.08-0.11)(0.03-0.06) (0.17-0.22) (0.17-0.22)
18.06 3.46 3.40 0.80 1.41 0.84 0.73 1.36 0.61 0.17 0.08 9.98 9.53
MoM 300.5 150.3
(17.59-18.55) (3.26-3.68) (3.20-3.62) (0.71-0.91) (1.28-1.55) (0.74-0.95) (0.64-0.83) (1.23-1.50) (0.52-0.70) (0.13-0.22)(0.06-0.12) (9.63-10.34) (9.05-10.03)
3.53 0.87 0.32 0.91 0.62 0.52 0.35 0.12 0.27 0.09 0.03 0.06 0.06
CoP 604.5 493.1
(3.38-3.68) (0.80-0.95) (0.27-0.36) (0.84-0.99) (0.56-0.69) (0.47-0.58) (0.30-0.40) (0.10-0.15) (0.24-0.32) (0.07-0.12)(0.02-0.05) (0.05-0.09) (0.04-0.08)
3.89 1.19 0.54 0.53 0.54 0.52 0.39 0.11 0.20 0.36 0.05 0.13 0.13
CoC 991.8 855.5
(3.77-4.01) (1.12-1.26) (0.49-0.58) (0.49-0.58) (0.49-0.58) (0.48-0.57) (0.36-0.43) (0.09-0.13) (0.18-0.23) (0.33-0.40)(0.03-0.06) (0.11-0.16) (0.11-0.15)
8.58 3.08 1.30 0.53 1.06 0.58 0.67 0.63 0.53 0.14 0.19 2.12 2.04
CoM 20.8 20.1
(7.40-9.93) (2.41-3.94) (0.89-1.90) (0.29-0.96) (0.70-1.61) (0.33-1.02) (0.40-1.14) (0.36-1.08) (0.29-0.96) (0.05-0.45)(0.07-0.51) (1.58-2.85) (1.50-2.77)
9.06 2.06 0.82 1.65 1.65 2.47 0.82 0.41 0.82 0.84
MoPoM 2.4 0 0 0 2.4
(5.97-13.76) (0.86-4.95) (0.21-3.29) (0.62-4.39) (0.62-4.39) (1.11-5.50) (0.21-3.29) (0.06-2.92) (0.21-3.29) (0.21-3.35)
*Including 36,765 with unknown fixation/bearing.

**Rates are likely to be underestimated: this reason was not solicited in the early phase of the registry (revision report forms MDSv1/MDSv2).
***For primaries from 2008 onwards the majority of revision report forms were MDSv3/MDSv6 which explicitly gave this indication for revision as an option.
101
102
Adverse reaction to
particulate debris
for primaries from
Pros- Number of
years Peripros- socket reaction to years per 1,000
Fixation/ at risk Aseptic Dislocation/ thetic Malalign- Implant Implant size particulate at risk prosthesis-
bearing type (x1,000) All causes loosening Pain Subluxation Infection fracture ment Lysis wear fracture mismatch debris** (x1,000) years
3.69 0.51 0.28 0.93 0.79 0.89 0.24 0.16 0.20 0.14 0.02 0.20 0.12
All hybrid 1,528.5 1,202.1
(3.60-3.79) (0.47-0.54) (0.26-0.31) (0.88-0.98) (0.75-0.84) (0.85-0.94) (0.21-0.26) (0.14-0.19) (0.18-0.22) (0.12-0.16) (0.01-0.03) (0.18-0.22) (0.10-0.14)
Hybrid and
3.60 0.53 0.23 1.02 0.77 0.97 0.24 0.17 0.23 0.09 0.02 0.07 0.06
MoP 910.2 700.2
(3.48-3.73) (0.48-0.58) (0.20-0.26) (0.95-1.09) (0.71-0.83) (0.91-1.04) (0.21-0.28) (0.14-0.20) (0.20-0.26) (0.07-0.11) (0.01-0.03) (0.05-0.09) (0.05-0.08)
15.97 2.98 2.83 1.25 1.17 1.81 0.49 1.62 0.34 0.34 0.08 7.48 6.38
MoM 26.5 10.0
(14.52-17.57) (2.39-3.72) (2.26-3.55) (0.89-1.75) (0.82-1.66) (1.37-2.40) (0.28-0.85) (1.20-2.19) (0.18-0.65) (0.18-0.65) (0.02-0.30) (6.50-8.59) (4.99-8.15)
3.46 0.26 0.15 0.98 0.98 0.82 0.18 0.08 0.15 0.10 0.02 0.04 0.03
CoP 355.3 324.8
(3.27-3.65) (0.21-0.32) (0.11-0.20) (0.89-1.09) (0.89-1.09) (0.73-0.92) (0.14-0.23) (0.05-0.11) (0.11-0.20) (0.07-0.14) (0.01-0.04) (0.02-0.06) (0.02-0.06)
2.89 0.51 0.42 0.41 0.49 0.52 0.29 0.11 0.15 0.38 0.02 0.13 0.13
CoC 225.4 156.2
(2.68-3.12) (0.43-0.62) (0.34-0.52) (0.34-0.51) (0.40-0.59) (0.44-0.63) (0.23-0.37) (0.07-0.16) (0.10-0.21) (0.31-0.47) (0.01-0.05) (0.09-0.19) (0.09-0.21)
6.21 0.72 1.08 2.03 1.79 0.12 0.12 0.36 0.12 0.12 0.12
MoPoM 8.4 0 0 8.4
(4.73-8.15) (0.32-1.60) (0.56-2.07) (1.26-3.27) (1.08-2.97) (0.02-0.85) (0.02-0.85) (0.12-1.11) (0.02-0.85) (0.02-0.85) (0.02-0.85)
5.11 0.57 1.14 1.14 2.84 0.57
CoPoM 1.8 0 0 0 0 0 0 1.8 0
(2.66-9.83) (0.08-4.03) (0.28-4.54) (0.28-4.54) (1.18-6.82) (0.08-4.03)
All reverse 3.90 1.27 0.31 0.84 0.73 0.65 0.28 0.20 0.23 0.05 0.02 0.10 0.07
193.4 169.1
hybrid (3.63-4.19) (1.12-1.44) (0.24-0.40) (0.72-0.98) (0.62-0.87) (0.54-0.77) (0.22-0.37) (0.14-0.27) (0.17-0.31) (0.03-0.10) (0.01-0.06) (0.07-0.16) (0.04-0.12)
Reverse hybrid and
3.89 1.20 0.21 0.89 0.74 0.72 0.25 0.21 0.22 0.05 0.08 0.06
MoP 128.8 0 112.9
(3.56-4.25) (1.03-1.41) (0.14-0.31) (0.74-1.07) (0.60-0.90) (0.59-0.89) (0.18-0.35) (0.14-0.31) (0.15-0.31) (0.02-0.10) (0.04-0.14) (0.03-0.13)
3.69 1.37 0.49 0.71 0.72 0.49 0.31 0.16 0.27 0.05 0.03 0.03
CoP 63.6 55.6 0
(3.25-4.20) (1.11-1.69) (0.34-0.69) (0.53-0.95) (0.54-0.97) (0.34-0.69) (0.20-0.49) (0.08-0.29) (0.17-0.43) (0.02-0.15) (0.01-0.13) (0.01-0.13)
All 10.38 2.19 3.06 0.25 0.46 1.10 0.57 0.90 0.25 0.23 0.06 3.73 3.04
435.5 182.1
resurfacing (10.08-10.68) (2.06-2.34) (2.90-3.23) (0.20-0.30) (0.40-0.53) (1.01-1.20) (0.51-0.65) (0.81-0.99) (0.20-0.30) (0.19-0.28) (0.04-0.08) (3.55-3.91) (2.80-3.31)
Resurfacing and
10.38 2.19 3.07 0.25 0.46 1.10 0.57 0.90 0.24 0.23 0.06 3.73 3.05
MoM 435.0 181.7
(10.08-10.69) (2.06-2.34) (2.91-3.24) (0.20-0.30) (0.40-0.53) (1.00-1.20) (0.51-0.65) (0.81-0.99) (0.20-0.29) (0.19-0.28) (0.04-0.09) (3.55-3.92) (2.81-3.31)
*Including 36,765 with unknown fixation/bearing.

Table 3.H10 PTIR estimates of indications for hip revision (95% CI) by years following primary hip replacement.
Adverse reaction to
particulate debris
for primaries from
Pros- Number of
Time years Peripros- socket reaction to years per 1,000
since at risk Aseptic Dislocation/ thetic Malalign- Implant Implant sizeparticulate at risk prosthesis-
primary (x1,000) All causes loosening Pain Subluxation Infection fracture ment Lysis wear fracture mismatch debris** (x1,000) years
4.59 1.13 0.60 0.80 0.69 0.70 0.31 0.27 0.25 0.14 0.03 0.70 0.45
All cases 8,159.6 5,913.6
(4.54-4.64) (1.10-1.15) (0.59-0.62) (0.78-0.82) (0.68-0.71) (0.68-0.72) (0.30-0.32) (0.26-0.28) (0.24-0.27) (0.14-0.15) (0.03-0.03) (0.68-0.72) (0.43-0.47)
8.24 0.99 0.50 2.47 1.97 1.70 0.70 0.07 0.30 0.20 0.10 0.09 0.10
<1 year 1,207.3 1,017.3
(8.08-8.41) (0.93-1.05) (0.46-0.54) (2.38-2.56) (1.89-2.05) (1.63-1.77) (0.66-0.75) (0.05-0.08) (0.27-0.34) (0.18-0.23) (0.08-0.12) (0.07-0.11) (0.09-0.12)
1 to 3 3.45 0.93 0.66 0.60 0.68 0.38 0.31 0.13 0.12 0.11 0.03 0.20 0.23
2,105.8 1,737.1
years (3.37-3.53) (0.89-0.97) (0.62-0.69) (0.56-0.63) (0.65-0.72) (0.35-0.41) (0.28-0.33) (0.11-0.14) (0.10-0.13) (0.10-0.13) (0.02-0.04) (0.18-0.22) (0.21-0.26)
3 to 5 3.40 0.89 0.71 0.44 0.42 0.44 0.22 0.19 0.17 0.10 0.02 0.65 0.51
1,651.7 1,301.7
years (3.31-3.49) (0.84-0.93) (0.67-0.75) (0.41-0.47) (0.39-0.45) (0.41-0.47) (0.20-0.25) (0.17-0.22) (0.15-0.19) (0.09-0.12) (0.01-0.02) (0.61-0.69) (0.47-0.55)
5 to 7 4.01 1.06 0.72 0.42 0.35 0.55 0.22 0.29 0.22 0.13 0.02 1.16 0.69
1,231.7 904.9
years (3.90-4.12) (1.01-1.12) (0.68-0.77) (0.38-0.46) (0.32-0.38) (0.51-0.59) (0.20-0.25) (0.26-0.32) (0.20-0.25) (0.12-0.16) (0.01-0.03) (1.10-1.22) (0.64-0.74)
7 to 10 4.58 1.39 0.53 0.50 0.35 0.64 0.21 0.48 0.31 0.16 0.01 1.35 0.84
1,187.1 747.3
years (4.46-4.70) (1.32-1.46) (0.49-0.58) (0.46-0.54) (0.32-0.39) (0.59-0.68) (0.19-0.24) (0.44-0.52) (0.28-0.35) (0.14-0.18) (0.01-0.02) (1.29-1.42) (0.78-0.91)
10 to 13 5.49 2.00 0.31 0.55 0.40 0.90 0.18 0.73 0.59 0.20 0.01 1.46 1.09
579.2 205.3
years (5.30-5.68) (1.88-2.11) (0.27-0.36) (0.50-0.62) (0.35-0.45) (0.82-0.98) (0.15-0.22) (0.66-0.80) (0.53-0.66) (0.17-0.24) (0.01-0.03) (1.36-1.56) (0.95-1.24)
13 to 15 5.57 2.28 0.29 0.60 0.39 0.85 0.17 0.94 0.92 0.25 1.30
150.4 0
years (5.21-5.96) (2.05-2.54) (0.22-0.39) (0.49-0.74) (0.30-0.50) (0.72-1.01) (0.12-0.25) (0.80-1.11) (0.78-1.09) (0.18-0.35) (1.13-1.50)
15 to 17 5.13 2.56 0.20 0.52 0.27 0.98 0.23 1.02 1.02 0.23 0.61
44.1 0
years (4.50-5.84) (2.13-3.08) (0.11-0.39) (0.35-0.78) (0.15-0.48) (0.72-1.31) (0.12-0.42) (0.76-1.37) (0.76-1.37) (0.12-0.42) (0.42-0.89)
4.98 2.72 1.36 1.36 0.45 1.36
≥17 years* 2.2 0 0 0 0 0 0
(2.76-8.99) (1.22-6.04) (0.44-4.21) (0.44-4.21) (0.06-3.21) (0.44-4.21)
*Current maximum observed follow-up is 17.75 years.

Note: Blank cells where there are no current data.
103
In Table 3.H10 (page 103), the PTIRs for each ceramic-on-polyethylene, ceramic-on-ceramic, hybrid
indication are shown separately for different time ceramic-on-ceramic and resurfacings, the PTIRs were
periods from the primary hip replacement, within reasonably consistent over time. In hybrid metal-on-
the first year, and between 1 to 3, 3 to 5, 5 to 7, 7 polyethylene and ceramic-on-polyethylene bearings,
to 10, 10 to 13, 13 to 15, 15 to 17, and ≥17 years there were marked increases at later time points. For
after surgery (the maximum follow-up for any implant pain, PTIRs were either fairly consistent or had a small
is now 17.75 years). Revision rates due to aseptic initial peak followed by a decline to fairly constant
loosening are fairly constant until five years and then rates for all bearings, apart from uncemented metal-
begin to steadily increase. Revision due to pain rises on-metal and resurfacings where rates started high,
out to seven years and then declines. The rates due to rose to peaks at five years and then declined again.
subluxation / dislocation, infection and malalignment Conversely, there was a high initial rate for dislocation
were all higher in the first year and then fell. In the case / subluxation in all fixation / bearing groups which later
of periprosthetic fracture, the highest rates were seen fell but then began to rise again in all groups apart
in the first year, these then declined markedly before from cemented metal-on-polyethylene, uncemented
beginning to rise again at around seven years. Adverse metal-on-metal, hybrid ceramic-on-ceramic and
reaction to particulate debris increased with time, as resurfacing (Figure 3.H11 (c)). Revision rates for
did lysis. infection were initially high and then fell in all groups
apart from uncemented metal-on-metal primary total
Figures 3.H11 (a) to 3.H11 (g) (pages 105 to 108) hip replacement and resurfacing (Figure 3.H11 (d)).
show how PTIRs for aseptic loosening, pain, The opposite was seen for lysis with increasing rates
dislocation / subluxation, infection, lysis and adverse over time in all groups (Figure 3.H11 (e)).
soft tissue reaction to particulate debris changed
with time. Only sub-groups with a total overall Revision rates due to an adverse reaction to
prosthesis-years at risk of more than 150,000 have particulate debris increased with time, up to seven
been included. With time from the operation, PTIRs for years in uncemented metal-on-metal primary total hip
aseptic loosening tended to rise in cemented fixations replacement and resurfacings (Figures 3.H11 (f) and
and follow a fairly similar pattern in uncemented (g)). Confidence Intervals have not been shown here
metal-on-polyethylene bearings. In uncemented metal- for simplicity but could be quite wide; these trends
on-metal, they rose and then fell. In uncemented require more in-depth investigation.
Figure 3.H11 (a) PTIR estimates of aseptic loosening by fixation and bearing.
0 to 1y
1 to 3y
(i) Cemented MoP

3 to 5y
5 to 7y
7 to 10y
10 to 13y
13 to 15y
≥15y
0 to 1y
1 to 3y
(ii) Cemented CoP

3 to 5y
5 to 7y
7 to 10y
10 to 13y
13 to 15y
≥15y
0 to 1y
1 to 3y
(iii) Uncemented MoP

3 to 5y
5 to 7y
7 to 10y
10 to 13y

13 to 15y
≥15y
0 to 1y
1 to 3y
(iv) Uncemented MoM

3 to 5y
5 to 7y
7 to 10y
10 to 13y
13 to 15y
≥15y
0 to 1y
1 to 3y
(v) Uncemented CoP

3 to 5y
5 to 7y
7 to 10y
10 to 13y
13 to 15y
≥15y
0 to 1y
1 to 3y
(vi) Uncemented CoC

3 to 5y
5 to 7y
7 to 10y
10 to 13y
13 to 15y
≥15y
0 to 1y
1 to 3y
(vii) Hybrid MoP

3 to 5y
5 to 7y
7 to 10y
10 to 13y
13 to 15y
≥15y
0 to 1y
1 to 3y
(viii) Hybrid CoP

3 to 5y
5 to 7y
7 to 10y
10 to 13y
13 to 15y
≥15y
0 to 1y
1 to 3y
(ix) Hybrid CoC

3 to 5y
5 to 7y
7 to 10y
10 to 13y
13 to 15y
≥15y
0 to 1y
1 to 3y
(x) Resurfacing
3 to 5y
5 to 7y
7 to 10y
10 to 13y
13 to 15y
≥15y
0 1 2 3 4 5 6 7 8
PTIR (per 1,000 prosthesis−years)
Figure 3.H11 (b) PTIR estimates of pain by fixation and bearing.
0 to 1y
1 to 3y
(i) Cemented MoP

3 to 5y
5 to 7y
7 to 10y
10 to 13y
13 to 15y
≥15y
0 to 1y
1 to 3y
(ii) Cemented CoP

3 to 5y
5 to 7y
7 to 10y
10 to 13y
13 to 15y
≥15y
0 to 1y
1 to 3y

3 to 5y
5 to 7y
7 to 10y
10 to 13y
13 to 15y
≥15y
0 to 1y
1 to 3y
(iv) Uncemented MoM

3 to 5y
5 to 7y
7 to 10y
10 to 13y
13 to 15y
≥15y
0 to 1y
1 to 3y
(v) Uncemented CoP

3 to 5y
5 to 7y
7 to 10y
10 to 13y
13 to 15y
≥15y
0 to 1y
1 to 3y
(vi) Uncemented CoC

3 to 5y
5 to 7y
7 to 10y
10 to 13y
13 to 15y
≥15y
0 to 1y
1 to 3y
(vii) Hybrid MoP

3 to 5y
5 to 7y
7 to 10y
10 to 13y
13 to 15y
≥15y
0 to 1y
1 to 3y
(viii) Hybrid CoP

3 to 5y
5 to 7y
7 to 10y
10 to 13y
13 to 15y
≥15y
0 to 1y
1 to 3y
(ix) Hybrid CoC

3 to 5y
5 to 7y
7 to 10y
10 to 13y
13 to 15y
≥15y
0 to 1y
1 to 3y
(x) Resurfacing
3 to 5y
5 to 7y
7 to 10y
10 to 13y
13 to 15y
≥15y
0 1 2 3 4 5 6 7 8
Figure 3.H11 (c) PTIR estimates of dislocation / subluxation by fixation and bearing.
0 to 1y
1 to 3y
(i) Cemented MoP

3 to 5y
5 to 7y
7 to 10y
10 to 13y
13 to 15y
≥15y
0 to 1y
1 to 3y
(ii) Cemented CoP

3 to 5y
5 to 7y
7 to 10y
10 to 13y
13 to 15y
≥15y
0 to 1y
1 to 3y

3 to 5y
5 to 7y
7 to 10y
10 to 13y
13 to 15y
≥15y
0 to 1y
1 to 3y
(iv) Uncemented MoM

3 to 5y
5 to 7y
7 to 10y
10 to 13y
13 to 15y
≥15y
0 to 1y
1 to 3y
(v) Uncemented CoP

3 to 5y
5 to 7y
7 to 10y
10 to 13y
13 to 15y
≥15y
0 to 1y
1 to 3y
(vi) Uncemented CoC

3 to 5y
5 to 7y
7 to 10y
10 to 13y
13 to 15y
≥15y
0 to 1y
1 to 3y
(vii) Hybrid MoP

3 to 5y
5 to 7y
7 to 10y
10 to 13y
13 to 15y
≥15y
0 to 1y
1 to 3y
(viii) Hybrid CoP

3 to 5y
5 to 7y
7 to 10y
10 to 13y
13 to 15y
≥15y
0 to 1y
1 to 3y
(ix) Hybrid CoC

3 to 5y
5 to 7y
7 to 10y
10 to 13y
13 to 15y
≥15y
0 to 1y
1 to 3y
(x) Resurfacing
3 to 5y
5 to 7y
7 to 10y
10 to 13y
13 to 15y
≥15y
0 1 2 3 4 5 6 7 8
Figure 3.H11 (d) PTIR estimates of infection by fixation and bearing.
0 to 1y
1 to 3y
(i) Cemented MoP

3 to 5y
5 to 7y
7 to 10y
10 to 13y
13 to 15y
≥15y
0 to 1y
1 to 3y
(ii) Cemented CoP

3 to 5y
5 to 7y
7 to 10y
10 to 13y
13 to 15y
≥15y
0 to 1y
1 to 3y

3 to 5y
5 to 7y
7 to 10y
10 to 13y
13 to 15y
≥15y
0 to 1y
1 to 3y
(iv) Uncemented MoM

3 to 5y
5 to 7y
7 to 10y
10 to 13y
13 to 15y
≥15y
0 to 1y
1 to 3y
(v) Uncemented CoP

3 to 5y
5 to 7y
7 to 10y
10 to 13y
13 to 15y
≥15y
0 to 1y
1 to 3y
(vi) Uncemented CoC

3 to 5y
5 to 7y
7 to 10y
10 to 13y
13 to 15y
≥15y
0 to 1y
1 to 3y
(vii) Hybrid MoP

3 to 5y
5 to 7y
7 to 10y
10 to 13y
13 to 15y
≥15y
0 to 1y
1 to 3y
(viii) Hybrid CoP

3 to 5y
5 to 7y
7 to 10y
10 to 13y
13 to 15y
≥15y
0 to 1y
1 to 3y
(ix) Hybrid CoC

3 to 5y
5 to 7y
7 to 10y
10 to 13y
13 to 15y
≥15y
0 to 1y
1 to 3y
(x) Resurfacing
3 to 5y
5 to 7y
7 to 10y
10 to 13y
13 to 15y
≥15y
0 1 2 3 4 5 6 7 8
Figure 3.H11 (e) PTIR estimates of lysis by fixation and bearing.
0 to 1y
1 to 3y
(i) Cemented MoP

3 to 5y
5 to 7y
7 to 10y
10 to 13y
13 to 15y
≥15y
0 to 1y
1 to 3y
(ii) Cemented CoP

3 to 5y
5 to 7y
7 to 10y
10 to 13y
13 to 15y
≥15y
0 to 1y
1 to 3y

3 to 5y
5 to 7y
7 to 10y
10 to 13y
13 to 15y
≥15y

0 to 1y
1 to 3y
(iv) Uncemented MoM

3 to 5y
5 to 7y
7 to 10y
10 to 13y
13 to 15y
≥15y
0 to 1y
1 to 3y
(v) Uncemented CoP

3 to 5y
5 to 7y
7 to 10y
10 to 13y
13 to 15y
≥15y
0 to 1y
1 to 3y
(vi) Uncemented CoC

3 to 5y
5 to 7y
7 to 10y
10 to 13y
13 to 15y
≥15y
0 to 1y
1 to 3y
(vii) Hybrid MoP

3 to 5y
5 to 7y
7 to 10y
10 to 13y
13 to 15y
≥15y
0 to 1y
1 to 3y
(viii) Hybrid CoP

3 to 5y
5 to 7y
7 to 10y
10 to 13y
13 to 15y
≥15y
0 to 1y
1 to 3y
(ix) Hybrid CoC

3 to 5y
5 to 7y
7 to 10y
10 to 13y
13 to 15y
≥15y
0 to 1y
1 to 3y
(x) Resurfacing
3 to 5y
5 to 7y
7 to 10y
10 to 13y
13 to 15y
≥15y
0 1 2 3 4 5 6 7 8
Figure 3.H11 (f) PTIR estimates of adverse soft tissue reaction by fixation and bearing.
0 to 1y
1 to 3y
(i) Cemented MoP

3 to 5y
5 to 7y
7 to 10y
10 to 13y
13 to 15y
≥15y
0 to 1y
1 to 3y
(ii) Cemented CoP

3 to 5y
5 to 7y
7 to 10y
10 to 13y
13 to 15y
≥15y
0 to 1y
1 to 3y

3 to 5y
5 to 7y
7 to 10y
10 to 13y
13 to 15y
≥15y
0 to 1y
1 to 3y
(iv) Uncemented MoM

3 to 5y
5 to 7y
7 to 10y
10 to 13y
13 to 15y
≥15y
0 to 1y
1 to 3y
(v) Uncemented CoP

3 to 5y
5 to 7y
7 to 10y
10 to 13y
13 to 15y
≥15y
0 to 1y
1 to 3y
(vi) Uncemented CoC

3 to 5y
5 to 7y
7 to 10y
10 to 13y
13 to 15y
≥15y
0 to 1y
1 to 3y
(vii) Hybrid MoP

3 to 5y
5 to 7y
7 to 10y
10 to 13y
13 to 15y
≥15y
0 to 1y
1 to 3y
(viii) Hybrid CoP

3 to 5y
5 to 7y
7 to 10y
10 to 13y
13 to 15y
≥15y
0 to 1y
1 to 3y
(ix) Hybrid CoC

3 to 5y
5 to 7y
7 to 10y
10 to 13y
13 to 15y
≥15y
0 to 1y
1 to 3y
(x) Resurfacing
3 to 5y
5 to 7y
7 to 10y
10 to 13y
13 to 15y
≥15y
0 5 10 15 20
Figure 3.H11 (g) PTIR estimates of adverse soft tissue reaction by fixation and bearing, since 2008.
0 to 1y
1 to 3y
(i) Cemented MoP 3 to 5y
5 to 7y
7 to 10y
10 to 13y
0 to 1y
1 to 3y
(ii) Cemented CoP 3 to 5y
5 to 7y
7 to 10y
10 to 13y
0 to 1y
1 to 3y
(iii) Uncemented MoP 3 to 5y
5 to 7y
7 to 10y
10 to 13y
0 to 1y
1 to 3y
(iv) Uncemented MoM 3 to 5y
5 to 7y
7 to 10y
10 to 13y
0 to 1y
1 to 3y
(v) Uncemented CoP 3 to 5y
5 to 7y
7 to 10y
10 to 13y
0 to 1y
1 to 3y
(vi) Uncemented CoC 3 to 5y
5 to 7y
7 to 10y
10 to 13y
0 to 1y
1 to 3y
(vii) Hybrid MoP 3 to 5y
5 to 7y
7 to 10y
10 to 13y
0 to 1y
1 to 3y
(viii) Hybrid CoP 3 to 5y
5 to 7y
7 to 10y
10 to 13y
0 to 1y
1 to 3y
(ix) Hybrid CoC 3 to 5y
5 to 7y
7 to 10y
10 to 13y
0 to 1y
1 to 3y
(x) Resurfacing 3 to 5y
5 to 7y
7 to 10y
10 to 13y
0 5 10 15 20
3.2.6 Mortality after primary hip the cumulative probability of death. While such surgery
may have contributed to the overall mortality, the
replacement surgery impact of this is not investigated in this report (see
In this section we describe the mortality of the survival analysis methods note in section 3.1). Among
cohort up to 15 years from primary hip replacement, the 1,251,164 primary hip replacements, there were
according to gender and age group. Deaths recorded 5,449 bilateral operations, with the left and right side
after 31 December 2020 were not included in the operated on the same day; here the second of the
analysis. For simplicity, we have not taken into two has been excluded, leaving 1,245,715 primary hip
account whether the patient had a first (or further) joint replacements, of whom 235,188 had died before the
revision after the primary operation when calculating end of 2020.
Table 3.H11 KM estimates of cumulative mortality (95% CI) by age and gender, in primary hip replacement.
Time since primary
Age group N 30 days 90 days 1 year 5 years 10 years 15 years

0.21 0.46 1.46 9.57 25.46 43.96
All cases 1,245,715*
(0.21-0.22) (0.45-0.48) (1.44-1.48) (9.52-9.63) (25.36-25.57) (43.77-44.16)
Males
0.07 0.16 0.53 2.32 5.28 9.73
<55 years 73,838
(0.06-0.10) (0.14-0.19) (0.48-0.58) (2.20-2.44) (5.07-5.49) (9.30-10.17)
0.06 0.20 0.62 3.37 8.83 17.01
55 to 59 years 51,033
(0.04-0.08) (0.16-0.24) (0.55-0.69) (3.20-3.55) (8.50-9.17) (16.35-17.70)
0.11 0.24 0.83 4.73 12.21 24.85
60 to 64 years 71,285
(0.09-0.14) (0.21-0.28) (0.76-0.90) (4.57-4.91) (11.90-12.54) (24.16-25.54)
0.16 0.35 1.10 6.85 18.90 38.74
65 to 69 years 84,761
(0.13-0.19) (0.32-0.40) (1.03-1.18) (6.66-7.04) (18.54-19.27) (37.99-39.49)
0.20 0.44 1.60 10.50 29.49 56.49

70 to 74 years 87,367
(0.18-0.24) (0.40-0.49) (1.52-1.68) (10.27-10.73) (29.06-29.92) (55.71-57.26)
0.38 0.75 2.49 16.68 46.45 77.97
75 to 79 years 70,989
(0.34-0.43) (0.69-0.82) (2.38-2.61) (16.37-16.99) (45.91-46.99) (77.13-78.80)
0.73 1.41 4.05 26.82 66.99 92.14
80 to 84 years 42,049
(0.65-0.82) (1.30-1.53) (3.86-4.25) (26.34-27.32) (66.28-67.70) (91.36-92.87)
1.61 2.93 7.63 43.37 86.32 98.29
≥85 years 18,174
(1.44-1.80) (2.70-3.19) (7.25-8.03) (42.53-44.22) (85.48-87.12) (97.67-98.77)
Females
0.06 0.20 0.64 2.50 5.14 8.54
<55 years 74,702
(0.05-0.08) (0.17-0.24) (0.59-0.70) (2.38-2.63) (4.94-5.36) (8.14-8.95)
0.06 0.18 0.58 2.99 7.15 13.11
55 to 59 years 59,053
(0.05-0.09) (0.15-0.22) (0.53-0.65) (2.84-3.15) (6.88-7.43) (12.58-13.67)
0.07 0.18 0.61 3.68 9.33 18.85
60 to 64 years 89,645
(0.05-0.09) (0.15-0.21) (0.56-0.66) (3.55-3.82) (9.08-9.59) (18.30-19.42)
0.08 0.21 0.74 4.83 13.74 29.16
65 to 69 years 123,960
(0.07-0.10) (0.19-0.24) (0.69-0.79) (4.70-4.97) (13.48-14.02) (28.58-29.74)
0.11 0.27 0.94 7.04 21.68 45.25
70 to 74 years 142,679
(0.10-0.13) (0.24-0.29) (0.89-0.99) (6.89-7.19) (21.37-21.99) (44.63-45.88)
0.21 0.44 1.47 11.41 34.58 66.53
75 to 79 years 126,850
(0.19-0.24) (0.40-0.47) (1.40-1.53) (11.21-11.61) (34.20-34.96) (65.86-67.19)
0.34 0.76 2.44 18.47 53.87 85.47
80 to 84 years 85,360
(0.30-0.38) (0.71-0.82) (2.34-2.55) (18.18-18.77) (53.38-54.37) (84.84-86.09)
0.80 1.74 4.76 32.28 75.25 95.76
≥85 years 43,970
(0.72-0.89) (1.62-1.87) (4.56-4.96) (31.78-32.78) (74.64-75.86) (95.19-96.27)
*Some patients had operations on the left and right side on the same day. The second of 5,449 pairs of simultaneous bilateral operations were excluded.
Table 3.H11 (page 109) shows Kaplan-Meier estimates fractured neck of femur compared to cases implanted
of cumulative percentage mortality at 30 days, 90 for other indications. A total of 45,144 (3.6%) of the
days and at 1, 5, 10 and 15 years from the primary hip primary total hip replacements were performed for a
replacement, for all cases and by age and gender. It is fractured neck of femur (NOF)†.
clear that younger patients had a lower risk of death.
These differences were apparent at 30 days, with Table 3.H12 shows that the proportion of primary
approximately half the risk of death for a male patient hip replacements performed for an indication of a
under the age of 55 compared to one aged 65 to 69 fractured neck of femur has continued to increase with
years. These differences persisted to one year and then time to a maximum of 7% in 2020, up from 5.2% in
diverged further with almost four times the risk of death 2017. The use of dual mobility bearings has become
in the older group at 15 years. For a similar age group more popular in this group and accounted for 8.4%
comparison, there was little initial difference for females of cases in 2020. The most striking feature is the
but by ten years, there was over twice the risk of death marked drop in 2020 in the total annual number of
in the older group. It is worthy of note that for all cases THRs performed for a fractured NOF (3,847 compared
in the registry, there is almost a 10% risk of death by to 5,383 in 2019). This is certainly due to the impact
five years, over 25% by ten years and over 40% by 15 of the COVID-19 pandemic, but how much is due to
years after primary hip replacement. fewer fractures occurring during lockdown and how
much is due to less provision of care because of the
3.2.7 Primary hip replacement for impact on services by the pandemic is not discernible
from these figures.
fractured neck of femur compared
with other reasons for implantation
Total hip replacement is an increasingly utilised
treatment option for fractured neck of femur and in this
section we report on revision and mortality rates for
primary total hip replacements performed as a result of
†These comprised 2,227 cases with the indication for primary hip replacement including fractured neck of femur in the early phase of the registry (i.e. 201,204
implants entered using MDSv1 and v2) and 42,917 cases with indications including acute trauma neck of femur in the later phase (i.e. 1,049,960 entered using
MDSv3, v6 and v7).
Table 3.H12 Number and percentage fractured NOF in the NJR by year.
NOF treated with

N (Primary total hip replacements for Dual mobility, Unipolar,
Year of primary all indications) N (NOF) (%) N(%) N(%)
2003 14,472 139 (1.0) 0 (0.0) 139 (100.0)
2004 28,106 292 (1.0) 0 (0.0) 292 (100.0)
2005 40,662 390 (1.0) 0 (0.0) 390 (100.0)
2006 48,511 528 (1.1) 0 (0.0) 528 (100.0)
2007 60,895 780 (1.3) 0 (0.0) 780 (100.0)

2008 67,434 867 (1.3) <4 (0.1) 866 (99.9)
2009 68,599 1,083 (1.6) 11 (1.0) 1,072 (99.0)
2010 71,119 1,370 (1.9) 8 (0.6) 1,362 (99.4)
2011 74,076 1,721 (2.3) 19 (1.1) 1,702 (98.9)
2012 78,282 2,439 (3.1) 21 (0.9) 2,418 (99.1)
2013 80,438 3,122 (3.9) 71 (2.3) 3,051 (97.7)
2014 87,682 3,725 (4.2) 144 (3.9) 3,581 (96.1)
2015 89,840 4,206 (4.7) 187 (4.4) 4,019 (95.6)
2016 94,346 4,872 (5.2) 292 (6.0) 4,580 (94.0)
2017 96,424 5,011 (5.2) 308 (6.1) 4,703 (93.9)
2018 96,771 5,369 (5.5) 328 (6.1) 5,041 (93.9)
2019 98,649 5,383 (5.5) 426 (7.9) 4,957 (92.1)
2020 54,858 3,847 (7.0) 324 (8.4) 3,523 (91.6)
Total 1,251,164 45,144 (3.6) 2,140 (4.7) 43,004 (95.3)
Table 3.H13 Fractured NOF vs. OA only by gender, age and fixation.
Reason for primary hip replacement

Fractured neck of femur Osteoarthritis only
(n=45,144) (n=1,102,840) Comparison
% Females 72.4% 59.2% P<0.001 (Chi-squared test)
Median age (IQR)
Both genders 73 (66 to 79) 70 (62 to 76) P<0.001 (Mann-Whitney U-test)

Males only 72 (65 to 79) 68 (60 to 75) P<0.001 (Mann-Whitney U-test)
Females only 73 (66 to 79) 71 (63 to 77) P<0.001 (Mann-Whitney U-test)
% Hip type*
All cemented 42.6 32.3
All uncemented 19.6 39.0
Overall P<0.001 (Chi-squared test)
All hybrid 35.5 22.6
All reverse hybrid 2.2 2.7
All resurfacing <0.1 3.4
*Excludes 103,180 cases who had other reasons in addition to osteoarthritis.
Table 3.H13 compares the fractured NOF group to 69, 70 to 74, 75 to 79, ≥80 for each gender). Figure
with the remainder with respect to gender and age 3.H12 (b) (page 114) shows similar cumulative revision
composition together and type of hip replacement rates for dual mobility compared to unipolar total hip
received. A significantly larger percentage of the replacement bearings in the hip fracture population out
fractured NOF cases, compared with the remainder, to five years at which point the numbers fall below 250
were women (72.4% versus 59.2%: P<0.001, Chi- in the dual mobility group. While the difference here
squared test). is not significant, it is interesting that this is a different
pattern seen to that for dual mobility bearings in
The fractured NOF cases were significantly older cemented and uncemented fixation groups in elective
(median age 73 years versus 70 years at operation: total hip replacement where the early revision rates
P<0.001 by Mann-Whitney U-test). We found that appear higher in the dual mobility bearings.
cemented and hybrid hips were used more commonly
in fractured NOF cases than in hip replacements Figure 3.H13 (page 115) shows a markedly worse
performed for other indications, but cemented fixation overall mortality in the fractured NOF cases compared
was still used in under half of the patients. Figure to cases implanted for other indications (P<0.001,
3.H12 (a) shows that the cumulative revision rate was logrank test). As in the overall mortality section, the
higher in the fractured NOF cases group compared second of 5,449 simultaneous bilateral procedures
with the remainder (P<0.001, logrank test). This effect were excluded. Gender and age differences did not
was not fully explained by differences in age and fully explain the difference seen, as a stratified analysis
gender, as stratification by these variables left the still showed a difference (P<0.001) but the results
result unchanged (P<0.001 using stratified logrank warrant further exploration.
test: 14 sub-groups of age <55, 55 to 59, 60 to 64, 65
Figure 3.H12 (a) KM estimates of cumulative revision for fractured NOF and OA only cases for primary
hip replacements. Blue italics in the numbers at risk table signify that fewer than 250 cases remained at risk
10

0
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
Years since primary
Osteoarthritis only 1,102,840 936,776 739,438 554,511 392,798 256,759 147,681 66,104 18,585
Fractured neck of femur 45,144 32,080 20,503 11,499 5,672 2,711 1,230 448 110
Figure 3.H12 (b) KM estimates of cumulative revision by bearing type for fractured NOF cases in
primary hip replacements. Blue italics in the numbers at risk table signify that fewer than 250 cases
remained at risk at these time points.
10
8
0
0 1 2 3 4 5 6 7 8 9 10
Years since primary
Unipolar bearings 43,004 30,871 19,958 11,336 5,648 2,706
Dual mobility 2,140 1,209 545 163 24 5
Figure 3.H13 KM estimates of cumulative mortality for fractured NOF and OA only in primary hip
replacements. Blue italics in the numbers at risk table signify that fewer than 250 cases remained at risk at
these time points.
70
60
50

Cumulative mortality (%)
40
30
20
10
0
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
Years since primary
Osteoarthritis only 1,098,083 942,744 748,882 566,282 405,672 269,415 158,076 71,406 20,125
Fractured neck of femur 45,115 32,581 20,953 11,828 5,910 2,875 1,331 493 124
3.2.8 Overview of hip revision The NJR asks surgeons and those responsible for
healthcare delivery to ensure that when primary and
procedures revision joint replacement procedures of the hip,
In this section we look at all hip revision procedures knee, ankle, elbow or shoulder are performed, that
performed since the start of the registry, 1 April 2003, the relevant MDS form is completed and data entered
up to 31 December 2020, for all patients with valid into the registry. This is a requirement mandated by
patient identifiers (i.e. whose data could therefore the Department of Health and Social Care. For the
be linked). purposes of the annual report, revision procedures
include any addition, removal or modification of the
In total, there were 129,308 revisions on 110,629 implants and procedures such as debridement and
individual patient sides (103,984 actual patients). implant retention with or without implant exchange,
In addition to the 37,444 first revised primary hip excision arthroplasty, amputation and conversion
replacements described in section 3.2.2 of this to arthrodesis. For the avoidance of confusion,
report, there were 84,275 revision procedures for completing a revision MDS form is also mandatory
which no primary hip replacement had been recorded for a procedure involving modification of a joint by
in the registry. adding another implant to another part of the joint.
For the analyses of surgeon performance, hospital
Revisions are classified as single-stage, stage one performance and implant performance, debridement
and stage two of two-stage revisions. Information on and implant retention without implant exchange is
stage one and stage two revisions are entered into currently excluded.
the registry separately, whereas in practice a stage
two revision has to be linked to a preceding stage
one revision. Although not all patients who undergo a
stage one of two revision will undergo a stage two of
two revision, in some cases stage one revisions have
been entered without a stage two, and vice versa,
making identification of individual revision episodes
difficult. We have made an attempt to do this later in
this section.
Table 3.H14 Number and percentage of hip revisions by procedure type and year.
Type of revision procedure

Year of revision Stage one of two- Stage two of two-
surgery Single-stage N(%) stage N(%) stage N(%) All procedures
2003* 16 (1.1) 0 (0.0) 1,418 (98.9) 1,434
2004 1,795 (65.8) 119 (4.4) 816 (29.9) 2,730
2005 3,460 (87.3) 202 (5.1) 303 (7.6) 3,965
2006 4,200 (86.8) 269 (5.6) 372 (7.7) 4,841
2007 5,563 (87.4) 339 (5.3) 460 (7.2) 6,362

2008 6,043 (86.2) 421 (6.0) 550 (7.8) 7,014
2009 6,319 (84.3) 517 (6.9) 661 (8.8) 7,497
2010 7,048 (86.5) 502 (6.2) 598 (7.3) 8,148
2011 7,978 (87.5) 530 (5.8) 611 (6.7) 9,119
2012 9,248 (88.1) 604 (5.8) 650 (6.2) 10,502
2013 8,536 (87.8) 567 (5.8) 623 (6.4) 9,726
2014 8,409 (87.0) 666 (6.9) 592 (6.1) 9,667
2015 8,018 (86.0) 709 (7.6) 597 (6.4) 9,324
2016 7,729 (87.3) 587 (6.6) 539 (6.1) 8,855
2017 7,697 (87.2) 614 (7.0) 517 (5.9) 8,828
2018 7,355 (87.7) 557 (6.6) 472 (5.6) 8,384
2019 7,076 (87.5) 546 (6.8) 465 (5.7) 8,087
2020 4,151 (86.0) 388 (8.0) 286 (5.9) 4,825
Total 110,641 (85.6) 8,137 (6.3) 10,530 (8.1) 129,308
*Incomplete year.
Note: Single-stages include DAIRs (Debridement And Implant Retention) and hip excision arthroplasty.
Table 3.H14 gives an overview of all hip replacement The incidence of revision hip replacement peaked in
revision procedures carried out each year since 2012 and has steadily declined since then, despite
April 2003. There were a maximum number of 11 the increasing number of at-risk implants prevailing in
documented revision procedures associated with the dataset.
any individual patient side (making up ten revision
episodes as one episode consisted of a stage one
of a two-stage procedure and a stage two of a two-
stage procedure).
Table 3.H15 (a) Number and percentage of hip revision by indication and procedure type.

Single-stage Stage one of two-stage Stage two of two-stage
Reason N(%) (n=110,641) N(%) (n=8,137) N(%) (n=10,530)
Aseptic loosening 51,525 (46.6) 954 (11.7) 2,270 (21.6)
Dislocation / Subluxation 18,205 (16.5) 326 (4.0) 536 (5.1)
Pain 18,188 (16.4) 819 (10.1) 908 (8.6)

Lysis 16,332 (14.8) 732 (9.0) 691 (6.6)
Implant wear 15,441 (14.0) 336 (4.1) 413 (3.9)
Periprosthetic fracture 13,292 (12.0) 323 (4.0) 466 (4.4)
Other indication 7,684 (6.9) 274 (3.4) 823 (7.8)
Malalignment 5,989 (5.4) 114 (1.4) 115 (1.1)
Infection 5,428 (4.9) 6,686 (82.2) 6,470 (61.4)
Implant fracture 4,017 (3.6) 83 (1.0) 162 (1.5)
Head-socket size mismatch 750 (0.7) 21 (0.3) 26 (0.2)
Adverse reaction to 10,217 (10.8) 238 (3.3) 169 (2.4)
particulate debris* n=94,586 n=7,132 n=7,096
*Not recorded in the early phase of the registry; MDSv3, v6 and v7 only.
Table 3.H15 (b) Number and percentage of hip revision by indication and procedure type in last five years.

Reason N(%) (n=34,008) N(%) (n=2,692) N(%) (n=2,279)
Aseptic loosening 13,080 (38.5) 203 (7.5) 167 (7.3)

Implant wear 4,595 (13.5) 99 (3.7) 55 (2.4)
Lysis 4,337 (12.8) 195 (7.2) 92 (4.0)
Adverse reaction to
3,722 (10.9) 108 (4.0) 65 (2.9)
particulate debris
Infection 2,632 (7.7) 2,349 (87.3) 1,823 (80.0)
Malalignment 1,603 (4.7) 29 (1.1) 14 (0.6)
Pain 1,535 (4.5) 46 (1.7) 35 (1.5)
Implant fracture 1,292 (3.8) 19 (0.7) 15 (0.7)
Head-socket size mismatch 134 (0.4) <4 (0.1) <4 (0.1)
Table 3.H15 (a) shows the stated indication for the five years is surgeons citing pain as an indication for
revision hip replacement surgery. Please note that, revision, falling from 16.4% to 4.5% of single-stage
as several indications can be stated, the indications revisions. There is also a higher proportion of cases
are not mutually exclusive and therefore column revised for periprosthetic fracture in the last five years
percentages may not add up to 100%. Aseptic (17.0% compared to 12.0%) and a higher proportion
loosening is the most common indication for revision. of cases revised due to infection (7.7% compared
to 4.9%). The ratio of stage two of two-stage, stage
Table 3.H15 (b) shows the stated indication for one of two-stage and single-stage revisions overall
revision hip replacement surgery performed in the last (1:0.77:10.5) is different compared to those performed
five years (1,826 days). The most notable difference, in the last five years (1:1.18:14.9).
between all the data and that recorded in the last
3.2.9 Rates of hip re-revision determined. For this purpose, an initial stage one
followed by either a stage one or a stage two have
In most instances (89.8% of 110,629 individual been considered to be the same revision episode
patient-sides), the first revision procedure was a and these were disregarded, looking instead for the
single-stage revision, however in the remaining start of a second revision episode. (The maximum
10.2% it was part of a two-stage procedure. For number of distinct revision episodes for any patient
a given patient side, survival following the first side was determined to be ten).
documented revision hip replacement procedure for
those with a linked primary in the registry (n=37,444) Kaplan-Meier estimates of the cumulative percentage
has been analysed. This analysis is restricted to probability of having a subsequent revision (re-revision)
patients with a linked primary procedure so that there were calculated. There were 4,253 re-revisions and,
is confidence that the next observed procedure on for 5,936 cases, the patient died without having been
the same joint is the first revision episode. If there is re-revised. The censoring date for the remainder was
no linked primary record in the dataset, it cannot be the end of 2020.
determined if the first observed revision is the first
Figure 3.H14 (a) plots Kaplan-Meier estimates of
revision or if it has been preceded by other revision
the cumulative probability of a subsequent revision
episodes. The time from the first documented
between 1 and 17 years since the primary operation.
revision procedure (of any type) to the time at which
a second revision episode was undertaken has been
Figure 3.H14 (a) KM estimates of cumulative re-revision in linked primary hip replacements (shaded area
indicates point-wise 95% CI). Blue italics in the numbers at risk table signify that fewer than 250 cases
25
20

Cumulative re−revision (%)
15
10
0
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
Years since first revision
Numbers at risk
37,444 27,757 20,187 13,780 8,428 3,688 1,451 376 65
Figure 3.H14 (b) KM estimates of cumulative re-revision by primary fixation in linked primary hip
these time points.
35
30
25
20
15
10
0
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
Cemented 8,564 5,863 3,906 2,442 1,426 679 307 86 18
Uncemented without MoM 10,935 7,936 5,571 3,577 2,090 970 367 106 17
Uncemented MoM 5,427 4,568 3,737 2,837 1,697 526 156 19 <4
Hybrid 5,647 3,757 2,429 1,448 845 407 157 49 10
Reverse hybrid 755 528 351 219 114 43 13 <4
Resurfacing 4,519 3,943 3,295 2,623 1,827 856 362 91 12
Figure 3.H14 (b) shows estimates of re-revision by

type of primary hip replacement. Resurfacing has the
lowest re-revision rate until approximately seven years,
after which the revision rate appears to be worse than
that associated with alternatives. However, after 12
years the numbers at risk are low and should therefore
be interpreted with caution.
Figure 3.H14 (c) shows the relationship between time and pain were more prevalent in the early period after
to first revision and the risk of subsequent revision. The the primary hip replacement and aseptic loosening and
earlier the primary hip replacement is revised, the higher pain later on. The relationship between (i) the time to
the risk of a second revision. There is a relationship first revision and the subsequent time to re-revision, and
between the indication for first revision and time to first (ii) the indication for the first revision and the time to re-
revision; earlier in this report (section 3.2.5) we show, revision requires further investigation.
for example, that revisions for dislocation / subluxation
Figure 3.H14 (c) KM estimates of cumulative re-revision by years to first revision, in linked primary hip
these time points.
25
20

15
10
0
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
First rev. <1y 9,951 7,287 5,312 3,698 2,514 1,485 749 255 60
First rev. 1 to 3y 7,256 5,689 4,499 3,405 2,476 1,381 573 119 5
First rev. 3 to 5y 5,617 4,504 3,669 2,917 1,949 620 125 <4
First rev. ≥5y 14,620 10,277 6,707 3,760 1,489 202 4
Figure 3.H15 (a) KM estimates of cumulative re-revision in cemented primary hip replacement by years
to first revision, in linked primary hip replacements. Blue italics in the numbers at risk table signify that
fewer than 250 cases remained at risk at these time points.
25
20
15
10
0
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
First rev. <1y 2,175 1,795 1,516 1,281 1,057 870 709 567 448 328 236 184 130 86 51 25 13 <4
First rev. 1 to 3y 1,783 1,544 1,349 1,174 1,018 872 724 584 476 355 277 194 139 79 33 12 5
First rev. 3 to 5y 1,175 986 837 703 597 499 406 335 255 177 121 74 37 15 <4
First rev. ≥5y 3,431 2,774 2,161 1,652 1,234 879 603 394 247 116 45 16 <4
For those with a documented primary hip replacement 1 to 3, 3 to 5 and more than 5 years. For cemented,
within the registry, Figures 3.H15 (a) to (e) show uncemented, hybrid, reverse hybrid and resurfacing hip
cumulative re-revision rates following the first revision replacements, those who had their first revision within
hip replacement, according to the main fixation used one year, or between one and three years of the initial
in the primary. Each sub-group has been further sub- primary hip replacement, experienced the worst re-
divided according to the time interval from the primary revision rates.
hip replacement to the first revision, i.e. less than 1 year,
Figure 3.H15 (b) KM estimates of cumulative re-revision in uncemented primary hip replacement by
years to first revision, in linked primary hip replacements. Blue italics in the numbers at risk table signify
that fewer than 250 cases remained at risk at these time points.
35
30
25

Cumulative re-revision (%)
20
15
10
0
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
First rev. <1y 00.04,406 3,839 3,354 2,929 2,550 2,202 1,800 1,486 1,211 908 666 480 296 162 86 47 18 5
First rev. 1 to 3y -1.03,208 2,897 2,592 2,350 2,094 1,864 1,608 1,377 1,150 893 598 347 194 97 39 13 0
First rev. 3 to 5y -2.02,553 2,320 2,116 1,931 1,763 1,603 1,423 1,212 904 498 191 84 33 11 0
First rev. ≥5y -3.06,195 5,293 4,442 3,598 2,901 2,221 1,583 999 522 166 41 7 0
Figure 3.H15 (c) KM estimates of cumulative re-revision in hybrid primary hip replacement by years to
first revision, in linked primary hip replacements. Blue italics in the numbers at risk table signify that fewer
than 250 cases remained at risk at these time points.
25
20
Cumulative re-revision (%)
15
10
0
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
First rev. <1y 00.02,201 1,798 1,468 1,189 924 731 541 405 324 247 186 131 85 56 32 19 10 <4
First rev. 1 to 3y -1.01,136 946 772 630 532 421 332 267 219 170 114 81 57 37 17 7 0
First rev. 3 to 5y -2.0 737 610 505 436 362 310 258 200 153 110 72 38 13 5 0
First rev. ≥5y -3.01,573 1,250 1,012 782 611 438 317 224 149 69 35 9 <4
Figure 3.H15 (d) KM estimates of cumulative re-revision in reverse hybrid primary hip replacement by
years to first revision, in linked primary hip replacements. Blue italics in the numbers at risk table signify
that fewer than 250 cases remained at risk at these time points.
30
25

20
15
10
0
0 1 2 3 4 5 6 7 8 9
First rev. <1y 273 237 211 178 146 113 95 71 51 36
First rev. 1 to 3y 175 160 129 115 99 81 64 51 39 26
First rev. 3 to 5y 103 91 78 65 49 40 32 17 13 9
First rev. ≥5y 204 149 110 75 57 41 28 19 11 6
Figure 3.H15 (e) KM estimates of cumulative re-revision in resurfacing primary hip replacement by years
to first revision, in linked primary hip replacements. Blue italics in the numbers at risk table signify that
35
30
25
20
15
10
0
0 1 2 3 4 5 6 7 8 9 10 11 12
First rev. <1y 478 459 437 412 390 373 346 329 314 290 268 226 176
First rev. 1 to 3y 684 652 627 599 577 556 531 514 480 418 317 221 151
First rev. 3 to 5y 833 804 785 766 743 700 669 633 537 372 205 100 34
First rev. ≥5y 2,524 2,338 2,094 1,838 1,585 1,310 1,077 784 496 205 66 16 <4
Table 3.H16 (a) shows the re-revision rate of the 37,444 the first year after surgery have approximately twice
primary hip replacements in the registry that were the chance of needing re-revision at each time point
revised. Of these, 4,253 were re-revised. Table 3.H16 compared with primaries that last more than five years.
(b) shows that primary hip replacements that fail within
Table 3.H16 (a) KM estimates of cumulative re-revision (95% CI).

Number of first Time since first revision
revised joints
at risk of
re-revision 1 year 3 years 5 years 10 years 15 years 17 years
Primary recorded 5.44 9.34 11.53 15.89 20.24 20.24
37,444
in the NJR (5.21-5.68) (9.03-9.66) (11.18-11.89) (15.37-16.42) (18.94-21.61) (18.94-21.61)
Table 3.H16 (b) KM estimates of cumulative re-revision (95% CI) by years since first failure.
Primary in the Time since first revision
NJR where the Number of first

first revision revised joints at
took place: risk of re-revision 1 year 3 years 5 years 7 years 10 years 13 years
<1 year after 7.59 12.53 14.85 17.25 20.05 23.13
9,951
primary (7.08-8.14) (11.86-13.23) (14.11-15.64) (16.40-18.13) (19.03-21.12) (21.72-24.62)
1 to 3 years 5.48 10.28 13.10 15.71 18.18 20.29
7,256
after primary (4.98-6.04) (9.57-11.03) (12.28-13.96) (14.78-16.69) (17.11-19.31) (18.81-21.86)
3 to 5 years 4.68 8.55 10.88 12.73 14.74 17.38
5,617
after primary (4.15-5.27) (7.82-9.34) (10.04-11.79) (11.79-13.74) (13.60-15.97) (15.24-19.78)
≥5 years after 4.23 6.86 8.39 9.37 10.80
14,620
primary (3.91-4.58) (6.43-7.31) (7.89-8.93) (8.79-9.98) (9.90-11.76)
Note: Maximum interval was 17.6 years.

Note: Data has not been presented at 15 years due to low numbers.
Table 3.H16 (c) KM estimates of cumulative re-revision (95% CI) by fixation and bearing used in primary
hip replacement. Blue italics signify that fewer than 250 cases remained at risk at these time points.
Bearing Time since first revision

Fixation surface N 1 year 3 years 5 years 7 years 10 years 13 years
5.44 9.34 11.53 13.51 15.89 18.69
All All 37,444
(5.21-5.68) (9.03-9.66) (11.18-11.89) (13.10-13.93) (15.37-16.42) (17.81-19.61)
6.08 9.52 11.48 13.25 15.90 17.17
All cemented 8,564
(5.59-6.62) (8.87-10.21) (10.74-12.28) (12.37-14.19) (14.73-17.16) (15.69-18.77)
6.08 9.36 11.21 13.04 15.65 16.56
MoP 7,674
(5.55-6.65) (8.69-10.08) (10.43-12.04) (12.12-14.02) (14.43-16.97) (15.12-18.12)
5.96 11.19 13.99 15.20 18.53 24.06
CoP 808
(4.49-7.88) (9.02-13.84) (11.40-17.12) (12.35-18.64) (14.55-23.44) (16.76-33.82)
All 5.28 9.58 11.66 13.61 15.63 17.65
16,362
uncemented (4.95-5.64) (9.11-10.06) (11.14-12.21) (13.01-14.24) (14.87-16.42) (16.44-18.95)
5.39 9.89 11.37 13.95 15.82 17.09
MoP 4,729
(4.77-6.09) (9.02-10.85) (10.40-12.42) (12.75-15.24) (14.32-17.45) (14.89-19.57)
4.63 8.60 10.85 12.77 14.79 16.62
MoM 5,427
(4.10-5.23) (7.87-9.40) (10.01-11.76) (11.83-13.79) (13.61-16.06) (14.88-18.55)
6.07 10.90 12.66 13.58 15.57 17.50
CoP 2,131
(5.12-7.20) (9.55-12.42) (11.15-14.37) (11.93-15.45) (13.39-18.08) (14.22-21.43)
5.49 9.72 12.15 13.94 16.05 18.91
CoC 3,855
(4.80-6.27) (8.78-10.75) (11.07-13.33) (12.72-15.27) (14.53-17.71) (16.37-21.79)
6.47 10.54 12.85 14.63 16.83 19.59
All hybrid 5,647
(5.84-7.16) (9.70-11.44) (11.88-13.90) (13.50-15.86) (15.37-18.42) (16.94-22.59)
6.64 10.23 12.32 13.80 15.69 17.44
MoP 3,277
(5.82-7.57) (9.17-11.41) (11.09-13.68) (12.40-15.35) (13.92-17.66) (14.76-20.56)
4.38 10.42 13.60 16.03 19.48 19.48
MoM 423
(2.78-6.87) (7.77-13.89) (10.48-17.55) (12.52-20.39) (15.11-24.91) (15.11-24.91)
6.83 10.92 14.12 15.14 15.14
CoP 1,228
(5.51-8.45) (9.12-13.06) (11.79-16.87) (12.50-18.29) (12.50-18.29)
5.97 10.69 12.14 14.93 18.09 27.12
CoC 652
(4.38-8.11) (8.45-13.47) (9.69-15.14) (11.92-18.61) (14.09-23.08) (17.92-39.77)
All reverse 5.64 9.12 10.14 12.01 15.03
755
hybrid (4.18-7.58) (7.17-11.57) (8.00-12.80) (9.45-15.20) (11.15-20.11)
5.88 9.21 9.98 12.94 16.52
MoP 501
(4.10-8.41) (6.85-12.33) (7.45-13.30) (9.60-17.32) (11.28-23.83)
All 3.21 6.60 9.31 11.49 14.61 19.99
4,519
resurfacing (2.73-3.77) (5.90-7.39) (8.45-10.25) (10.51-12.56) (13.37-15.96) (17.70-22.52)
6.38 9.69 12.14 15.08 16.99 17.97
Unclassified 1,597
(5.26-7.72) (8.27-11.34) (10.49-14.03) (13.13-17.29) (14.74-19.53) (15.15-21.26)
Note: Maximum interval was 17.6 years.

Note: Data has not been presented at 15 years due to low numbers.
Table 3.H16 (c) shows cumulative re-revision rates at The failure rates for revisions following resurfacings were
1, 3, 5, 7, 10 and 13 years following the first revision comparatively low, but Figure 3.H14 (b) (page 120)
for those with documented primary hip replacements shows that after ten years the failure rate is becoming
within the registry, broken down by fixation types and higher than those for alternatives.
bearing surfaces used in the primary hip replacement.
3.2.10 Reasons for hip re-revision column in Table 3.H17 (b), we report the indications
for all the second linked revisions e.g. 888 linked
Tables 3.H17 (a) and (b) show a breakdown of the second revisions recorded aseptic loosening as an
stated indications for the first revision and for any indication. It is interesting to note that both dislocation
second revision (please note the indications are and infection are much more common indications for a
not mutually exclusive). Table 3.H17 (a) shows the second revision than for a first revision. This shows the
indications for recorded revisions in the registry and increased risk of instability and infection following the
Table 3.H17 (b) reports the indications for the first linked first revision of a hip replacement compared to that of
revision and the number and percentage of first linked primary hip replacement.
revisions that were subsequently revised. In the final
Table 3.H17 (a) Number of revisions by indication for all revisions.
Reason for revision All recorded revisions, N(%)

Aseptic loosening 54,749 (42.3)
Pain 19,915 (15.4)
Dislocation / Subluxation 19,067 (14.7)

Infection 18,584 (14.4)
Lysis 17,755 (13.7)
Implant wear 16,190 (12.5)
Periprosthetic fracture 14,081 (10.9)
Malalignment 6,218 (4.8)
Implant fracture 4,262 (3.3)
Head/socket size mismatch 797 (0.6)
Other indication 8,781 (6.8)
Adverse reaction to particulate debris* 10,624 (8.2)
*Adverse reaction to particulate debris was only recorded using MDSv3 onwards and as such was only a potential reason for revision among a total of 108,814
revisions as opposed to 129,308 revisions for the other reasons.
Table 3.H17 (b) Number of revisions by indication for first linked revision and second linked re-revision.
First linked revision Second linked revision

Subsequently
Reason for revision N re-revised, N(%) N
Aseptic loosening 9,190 874 (9.5) 888
Dislocation / Subluxation 6,503 762 (11.7) 1,057
Periprosthetic fracture 5,696 589 (10.3) 351

Infection 5,660 991 (17.5) 1,343
Pain 4,916 620 (12.6) 414
Malalignment 2,524 242 (9.6) 204
Lysis 2,209 181 (8.2) 171
Implant wear 2,078 187 (9.0) 205
Implant fracture 1,167 127 (10.9) 121
Head/socket size mismatch 245 37 (15.1) 15
Other indication 3,098 416 (13.4) 269
Adverse reaction to particulate debris* 2,575 246 (9.6) 121
*Adverse reaction to particulate debris was only recorded using MDSv3 onwards and as such was only a potential reason for revision among a total of 23,977
revisions as opposed to 37,444 revisions for the other reasons.
Tables 3.H18 (a) and (b) show that the numbers of improved data capture over time, improved linkability
revisions and the relative proportion of revisions with of records and the longevity of hip replacements with a
a linked primary in the registry increased with time. proportion of primaries being revised being performed
Approximately 57% of revisions performed in 2020 had before data capture began or being outside the
a linked primary in the registry. This is likely to reflect coverage of the registry.
Table 3.H18 (a) Number of revisions by year.

Number of first revisions (%) with the
Year of first revision in the NJR* Number of first revisions* associated primary recorded in the NJR
2003 1,411 43 (3.0)
2004 2,639 142 (5.4)
2005 3,748 304 (8.1)
2006 4,482 462 (10.3)
2007 5,858 811 (13.8)
2008 6,314 1,155 (18.3)

2009 6,561 1,512 (23.0)
2010 7,073 1,949 (27.6)
2011 7,945 2,655 (33.4)
2012 9,026 3,338 (37.0)
2013 8,224 3,041 (37.0)
2014 8,085 3,092 (38.2)
2015 7,655 3,231 (42.2)
2016 7,273 3,227 (44.4)
2017 7,178 3,333 (46.4)
2018 6,817 3,466 (50.8)
2019 6,518 3,489 (53.5)
2020 3,822 2,194 (57.4)
Total 110,629 37,444 (33.8)
*First documented revision in the NJR.
Table 3.H18 (b) Number of revisions by year, stage, and whether or not primary is in the NJR.
Year of first Single-stage First documented stage of two-stage

revision in the
NJR* Primary not in the NJR Primary in the NJR Primary not in the NJR Primary in the NJR
2003 16 0 1,352 43
2004 1,668 94 829 48
2005 3,117 249 327 55
2006 3,645 374 375 88
2007 4,603 683 444 128

2008 4,688 954 471 201
2009 4,571 1,250 478 262
2010 4,712 1,718 412 231
2011 4,899 2,387 391 268
2012 5,310 3,011 378 327
2013 4,869 2,742 314 299
2014 4,644 2,798 349 294
2015 4,116 2,907 308 324
2016 3,810 2,940 236 287
2017 3,598 3,059 247 274
2018 3,126 3,217 225 249
2019 2,852 3,220 177 269
2020 1,497 1,997 131 197
Total 65,741 33,600 7,444 3,844
3.2.11 90-day mortality after reflect the fact that this patients in this group were
younger at the time of their first revision, median age
hip revision of 69 (IQR 61 to 77) years compared to the group
The overall cumulative percentage mortality at 90 days without primaries documented in the registry who
after hip revision was lower in the cases with a primary had a median age of 74 (IQR 66 to 80) years. The
hip replacement recorded in the registry compared percentage of males to females was similar in both
with the remainder (Kaplan-Meier estimates 1.42 (95% groups (44.2% versus 42.5% respectively).
CI 1.30-1.54) versus 1.90 (1.80- 2.00)), which may
3.2.12 Conclusions most utilised construct in 2020 (19.4% of all THRs),
followed by cemented metal-on-polyethylene (17.3%)
As in previous annual reports, our analysis of implants and then uncemented ceramic-on-polyethylene
has been by revision of the construct, rather than (17.1%). This is our second year reporting on dual
revision of a single component, as the mechanisms of mobility; this is used in different bearing combinations
failure (such as wear, adverse reaction to particulate and the numbers this year enable us to report on
debris and dislocation) are interdependent between metal-on-polyethylene-on-metal and ceramic-on-
different parts of the construct. Revision analyses polyethylene-on-metal within some sub-groups and
have also been stratified by age and gender. The their use does seem to be increasing. Given that the
highest failure rates are among younger women proposed benefits of dual mobility bearings include
and the lowest among older women. When data on reduced risk of early revision due to dislocation,
metal-on-metal is excluded, younger women have perhaps at an increased risk of long-term wear, it
similar revision rates to younger men. Once again, it is interesting to note that for elective indications,
must be emphasised that implant survivorship is only there appears to be a higher risk of early revision.
one measure of success and cannot be used as an The numbers are not yet sufficient to comment on
indication of satisfaction, relief of pain, improvement longer-term risks or the sub-groups described. It is
in function and the resulting greater participation in possible that this is a case mix selection effect and our
society. The data clearly show that constructs failing at annual report will continue to report on these patterns,
different rates is associated with the age and gender particularly if adoption continues to increase. We
of the recipients. observed a different pattern when dual mobility is used
for patients with a fractured neck of femur without this
Overall, the number of primary hip replacements early higher rate of revision.
recorded annually in the registry continues to increase
with 1,369,888 now recorded, of which 1,251,164 Since the 12th NJR Annual Report in 2015, our data
were eligible for analysis. The COVID-19 pandemic has been presented by age and gender comparing
has had a marked impact on the provision of hip combinations of fixation and bearing. This assists
replacement with primary THR decreasing from clinicians and patients in choosing classes of
98,649 in 2019 to 54,858 in 2020 and revision THR prostheses that are the most appropriate for particular
from 8,087 in 2019 to 4,825 in 2020. This represents types of patients. For example, in males aged 55
a massive under-provision with significant implications to 64 years, at 15 years post-surgery, hybrid and
for morbidity among patients and we examine this uncemented ceramic-on-polyethylene and ceramic-
impact further in our COVID-19 section of the report on-ceramic constructs as well as cemented ceramic-
(see page 341). Similarly, the provision of THR for a on-polyethylene constructs have similarly low revision
fractured neck of femur has decreased markedly from rates of approximately 5%, while cemented metal-
5,383 in 2019 to 3,847 in 2020. on-polyethylene constructs have revision rates of
8.65% (95% CI 7.82-9.57) and uncemented metal-
Since 2003 the types of implants utilised have on-polyethylene bearings 8.13% (95% CI 7.08-9.32).
changed dramatically and these changes continue. Resurfacings in this group have an even higher revision
Between 2003 and 2007 cemented fixation was the rate at 15 years of 10.49% (95% CI 9.69-11.35).
most common, followed by uncemented fixation. Women aged 55 to 64 years have lower revision rates
Between 2008 and 2019 uncemented fixation was the than men for all fixation/bearing combinations at 15
most common, with hybrid fixation increasing steadily years, except for those with metal-on-metal bearings,
from 2012 to become the most commonly used such as resurfacings, where the revision rates are
fixation for the first time in 2020. markedly higher for women than men and markedly
higher than alternatives. For example, 15-year revision
Since 2011, the use of ceramic-on-ceramic bearings
rates with hybrid ceramic-on-ceramic constructs in
has declined while the use of ceramic-on-polyethylene
this group are 3.03% (95% CI 2.54-3.62) compared
bearings has increased markedly, with ceramic-on-
to metal-on-metal hip resurfacing of 22.56% (95% CI
polyethylene hybrid total hip replacements being the
21.22-23.97).
For patients over 75 years, all combinations except bearing type for commonly used uncemented implants
those with metal-on-metal bearings have good shows that results are acceptable if metal-on-metal
outcomes, with cemented and hybrid ceramic-on- bearings are excluded. It is important to note that
polyethylene possibly having the lowest failure rates. there is variability in brand level constructs with
The risk of revision at 15 years in this group is very variation in outcome according to factors such as the
small, males 5.01% (95% CI 4.63-5.43) and females bearing combination used. It is therefore important
3.52% (95% CI 3.32-3.73). The 15-year mortality rate to consider the construct when selecting implants for
in men aged 75 to 79 years is 77.97% (95% CI 77.13- specific outcomes. We encourage all readers to view
78.80) and in women aged 75 to 79 years is 66.53% Table 3.H8 for fine details of construct performance.
(95% CI 65.86-67.19). This clearly shows that in older
patients, the vast majority of treatment strategies will Metal-on-metal stemmed and resurfacing implants
last the rest of the patients’ lives. Even in those aged continue to fail at higher than expected rates and their
65 to 69 years at the time of surgery, only 61% of use is now extremely rare. The best performing brand of
males and 71% of females are still alive 15 years later. resurfacing has a failure rate of 11.30% (95% CI 10.74-
11.88) at 17 years. The use of metal-on-metal bearings
We have also examined outcomes of different head has undoubtedly led to a large excess of revisions
sizes (bearing diameters) with alternative fixation and which would not have occurred if alternate bearings
bearing types and these results are interesting. With had been used. This has been modelled and published
metal-on-polyethylene and ceramic-on-polyethylene, in the Journal of Bone and Joint Surgery. For every 100
large head sizes appear to be associated with higher MoM hip-resurfacing procedures, it is estimated that
failure rates particularly with 36mm heads used with there would be 7.8 excess revisions by ten years, and
cemented fixation and heads >36mm used with similarly for every 100 stemmed MoM THR procedures
uncemented fixation. Ceramic-on-ceramic bearings that there would be 15.9, which equates to 8,021
have lower failure rates with larger bearings when excess first revisions (Hunt et al., 2018).
used with uncemented fixation, as predicted by Alison
Smith’s flexible parametric survival models published It is striking to note the high rates of revision for adverse
in the Lancet in 2012 (Smith et al., 2012). However, soft tissue reaction to particulate debris in patients who
this does not appear to hold true with ceramic-on- have received metal-on-metal bearings. Analysis of
ceramic hybrid fixations. This demonstrates how stemmed metal-on-metal bearings by head size shows
important it is to examine the entire construct, not just that 28mm heads have the best survivorship, but this is
the individual variables such as fixation, composition of still poor compared to alternatives.
bearing and head size.
We note that revision rates by year of surgery for
With regard to specific branded stem / cup the entire cohort increased dramatically from 2003
combinations, some of the best implant survivorships to 2008 and then began to decline and continue
have still been found to be achieved by mix and match to do so. The peak rate matches that for the use
cemented hard-on-soft bearing constructs, although of resurfacing arthroplasty and stemmed metal-
this practice remains contrary to both the MHRA and on-metal, with the peak usage of these devices in
implant manufacturers’ guidelines for usage. 2008 corresponding with the highest failure rates
by year of primary surgery. This demonstrates the
It is encouraging that the most commonly used profoundly negative effect metal-on-metal has had
constructs by brand in cemented and hybrid on hip replacement outcomes. However, as this
fixation have good results. This does not hold true temporal trend is also present after knee replacement,
for uncemented fixation, but further breakdown by although with a lesser magnitude, it is likely that
Smith AJ, Dieppe P, Vernon K, Porter M, Blom AW; National Joint Registry of England and Wales. Failure rates of stemmed metal-on-metal hip replacements:
analysis of data from the National Joint Registry of England and Wales. Lancet. 2012 Mar 31;379(9822):1199-204.
Hunt LP, Whitehouse MR, Beswick A, Porter ML, Howard P, Blom AW; Implications of Introducing New Technology: Comparative Survivorship Modelling of Metal-
on-Metal Hip Replacements and Contemporary Alternatives in the National Joint Registry. J Bone Joint Surg Am. 2018 Feb 7;100(3):189-196.
other factors also contribute to the decline in revision Risk of re-revision rate is strongly associated with time
rates. For example, the decline coincides with the to first revision; 20.05% (95% CI 19.03-21.12) of hips
commencement of the NJR’s clinician feedback revised within a year of primary surgery are re-revised
activity. It is noteworthy that this decline appears to be within ten years. In contrast, when the primary lasts at
ongoing, which is undoubtedly very good news. least five years the re-revision rate is 10.80% (95% CI
9.90-11.76) at ten years. Re-revision rates up to ten
Consistent with results from previous years’ annual years appear to be independent of the fixation and
reports, we observed similar revision rates for total hip bearing of the primary hip replacement, except for
replacement performed as a result of a fractured neck resurfacing procedures which are initially associated
of femur and those carried out for other causes. As with lower re-revision rates, but this pattern appears to
expected, mortality rates were higher for the fractured wane between seven and ten years after the re-revision.
neck of femur group.
Overall, this report is good news for patients,
The number of revision total hip replacements clinicians and the healthcare sector. Provision of hip
recorded in the registry increased to a peak of 10,502 replacement increased up to 2019, revision rates
in 2012 and since then has declined steadily to 8,087 continued to decline and clinicians are increasingly
in 2019, with a marked drop to 4,825 in 2020 due utilising constructs with proven longevity. In contrast,
to the impact of the COVID-19 pandemic. Please in 2020 there was a massive under-provision of both
note that there may be late registrations for 2020 primary and revision hip replacement with over 47,000
procedures and thus the figure for this year may be fewer hip replacements performed than in 2019.
revised upwards in the next annual report. Aseptic As hip replacement is undertaken to treat severe
loosening is the most common reason for revision, pain and functional limitation, this deficit represents
accounting for nearly half of all cases, followed by pain considerable suffering for a large cohort of people
and instability. nationally. (See our patient perspective in the COVID
section of the report on page 341).
3.3 Outcomes after
knee replacement
3.3.1 Overview of primary knee past. This distinction is available for cases reported
on the MDSv7 forms but not in previous versions.
replacement surgery Cases are therefore not reported separately in this
In this section of the report we address revision and year’s report, but work is ongoing to determine if
mortality outcomes for all primary knee operations this distinction can be defined from data entered
performed between 1 April 2003 and 31 December in previous versions of the MDS forms with the
2020. The very first patients who were entered into introduction of our new component database. If this
the registry therefore had a potential 17.75 years of is possible, it will be reported in future annual reports.
follow-up. The term multicompartmental knee replacement has
been introduced to refer to instances when more
The outcomes of total and partial knee replacement than one unicompartmental construct is implanted
procedures are discussed throughout this simultaneously i.e. one patellofemoral and one
section, hereafter referred to as total (TKR) and unicondylar, or one patellofemoral and two unicondylar.
unicompartmental (UKR) knee replacement. Brief details
of the type of orthopaedic surgery involved for each Figure 3.K1 (a) describes the data cleaning processes
form of replacement can be found in section 3.1. We applied to produce the total of 1,357,077 primary
note here that the NJR data collection process now knee procedures included in the analyses we present
distinguishes between medial and lateral unicondylar in this section.
replacements, although this was not the case in the
National Joint Registry | 18th Annual Report | Knees
Figure 3.K1 (a) Knee cohort flow diagram.
Knee procedures recorded by the NJR

N=1,564,300
*Reoperation procedures N=1,530

*Non−consenting procedures N=53,598
*Non−traced procedures N=44,513
*Invalid IDs N=0

N=1,465,011
*Procedures prior to April 2003 N=41


N=1,464,930


N=1,464,917
Northern Ireland N=13,301

Isle of Man N=648


N=1,450,968

/ Prostheses used / Indications / Unit N=1,323
Duplicate same day revision procedures based on:
NHS No. / Date / Side / Procedure type N=45
Unique procedures
N=1,449,600
Procedures (2,341 knees) with

an inconsistent operative pattern N=4,988
Procedures (1,389,119 knees)

N=1,444,612
All revision procedures N=87,535

*Of which, knee procedures where the first recorded
procedure in a sequence is a revision N=37,816
Primary procedures
(Revision analyses)
N=1,357,077
Bilateral procedures (same day) N=13,510
N=1,343,567 * Reasons not necessarily mutually exclusive
Over the lifetime of the registry, the 1,357,077 primary Table 3.K1 shows the breakdown of cases by type of
knee joint replacement procedures contributing to our knee replacement, the method of fixation, constraint
revision analyses were carried out by a total of 3,446 and bearing used. A breakdown within each method
unique consultant surgeons working across 469 units. of fixation of the percentage of constraint and bearing
types used is shown in a separate column. Cemented
Over the last three years (1 January 2018 to 31 TKR is the most commonly performed type of knee
December 2020), 260,620 primary knee procedures replacement (83.7% of all primary knee replacements).
(representing 19.2% of the current registry) were A further 4.2% were either all uncemented or hybrid
performed by 1,864 consultant surgeons working TKRs. Most unicompartmental knee replacements
across 406 units. Looking at caseload over this three were unicondylar (9.3% of the total) with the remainder
year period, the median number of primary procedures being patellofemoral (1.2%).
per consultant surgeon was 107 (IQR 39 to 198)
and the median number of procedures per unit was More than half of all operations (57.6%) were TKRs
576 (IQR 284 to 860). A proportion of surgeons will which were all cemented and unconstrained (cruciate
have commenced practice as a consultant during retaining) with a fixed bearing, followed by 19.9%
this period, some may have retired, and some which were all cemented and posterior stabilised with
surgeons may have periods of surgical inactivity within a fixed bearing. Within each method of fixation, it can
the coverage of the NJR, therefore their apparent be seen that uncemented and hybrid prostheses are
caseload would be lower. mostly unconstrained. However, while uncemented
are almost equally likely to have a mobile or fixed
Over this three year period, there have been 226,350 bearing, hybrid knees are more likely to utilise a
primary total knee replacements performed by 1,852 fixed bearing. Approximately two-thirds (68.8%) of
surgeons (median=96 cases per surgeon; IQR 37 cemented TKRs are unconstrained and have a fixed
to 171) in 403 separate units (median=580 cases bearing. Unicondylar knee surgery typically involves
per unit; IQR 298 to 877). In the same time period, the use of a mobile bearing (61.5%). A number of
there have been 30,068 primary unicondylar knee primary knee replacements could not be classified
procedures performed by 794 consultant surgeons according to their bearing / constraint (approximately
(median=19 cases per surgeon; IQR 5 to 50) in 364 1.6% of the total cohort).
units (median=49 cases per unit; IQR 17 to 103).
The majority of primary knee replacements were

carried out on women (females 56.3%; males 43.7%).
The median age at primary operation was 70 years
(IQR 63 to 76) and the overall range was 7 to 100
years, see Table 3.K3 (page 146) and commentary
later for discussion of age at primary by type of knee
replacement. Osteoarthritis was given as a documented
indication for surgery in 1,321,874 procedures (97.4%
of the cohort) and was the sole indication given in
1,310,663 (96.6%) primary knee procedures.
Table 3.K1 Number and percentage of primary knee replacements by fixation, constraint and bearing.
Type of primary knee operation Percentage of each

constraint type used Percentage of
Number of primary within each method all primary knee
Fixation method Constraint and bearing type knee operations of fixation operations
All types 1,357,077 100.0
Total knee replacement
All cemented 1,136,212 83.7
Cemented and unconstrained, fixed 781,402 68.8 57.6
unconstrained, mobile 40,231 3.5 3.0
posterior-stabilised, fixed 270,635 23.8 19.9
posterior-stabilised, mobile 12,886 1.1 0.9
constrained condylar 10,698 0.9 0.8
monobloc polyethylene tibia 18,296 1.6 1.3
pre-assembled/hinged/linked 2,064 0.2 0.2
All uncemented 47,061 3.5
Uncemented and unconstrained, fixed 18,187 38.6 1.3

posterior-stabilised, fixed 3,428 7.3 0.3
other constraints 294 0.6 <0.1
All hybrid 9,851 0.7
Hybrid and unconstrained, fixed 6,468 65.7 0.5
posterior-stabilised, fixed 819 8.3 0.1
other constraints 408 4.1 <0.1
Unicompartmental knee replacement
All unicondylar,
96,187 7.1
cemented
Cemented and fixed 40,281 41.9 3.0
mobile 49,610 51.6 3.7
monobloc polyethylene tibia 6,296 6.5 0.5
All unicondylar,
29,268 2.2
uncemented/hybrid
Uncemented/hybrid
fixed 1,259 4.3 0.1
and
mobile 27,606 94.3 2.0
monobloc polyethylene tibia 403 1.4 <0.1
Patellofemoral 15,639 1.2
Multicompartmental 586 <0.1
Unclassified 22,273 1.6
Figure 3.K1 (b) Frequency of primary TKR within elective cases stratified by procedure type. Consultants
have been placed in groups by the volume of cases they undertake per annum. Each colour represents
total volume of cases undertaken by all the consultants in that grouping.
100,000
75,000
Frequency (N=)
50,000
25,000
0
2003 2005 2007 2009 2011 2013 2015 2017 2019
2004 2006 2008 2010 2012 2014 2016 2018 2020
N =Procedures per year
1≤N≤2 3≤N≤4 5≤N≤6 7≤N≤12 13≤N≤24 25≤N≤48 49≤N≤96 ≥97
Figure 3.K1 (c) Frequency of primary UKR within elective cases stratified by procedure type.
Consultants have been placed in groups by the volume of cases they undertake per annum.
Each colour represents total volume of cases undertaken by all the consultants in that grouping.
12,500
10,000

Frequency (N=)
7,500
5,000
2,500
0
2003 2005 2007 2009 2011 2013 2015 2017 2019
2004 2006 2008 2010 2012 2014 2016 2018 2020
1≤N≤2 3≤N≤4 5≤N≤6 7≤N≤12 13≤N≤24 25≤N≤48 49≤N≤96 ≥97
Figure 3.K1 (d) Frequency of primary patellofemoral knee replacements within elective cases stratified
by procedure type. Consultants have been placed in groups by the volume of cases they undertake per
annum. Each colour represents total volume of cases undertaken by all the consultants in that grouping.
1,000
750
Frequency (N=)
500
250
0
2003 2005 2007 2009 2011 2013 2015 2017 2019
2004 2006 2008 2010 2012 2014 2016 2018 2020
1≤N≤2 3≤N≤4 5≤N≤6 7≤N≤12 13≤N≤24 25≤N≤48 49≤N≤96
Figures 3.K1 (b) to (d) show the yearly number data collection started. Prior to 2020 the majority of
of primary knee replacements performed for all additional procedures were contributed by higher
indications. Procedures have been stratified by volume surgeons i.e. those performing over 49
total knee, unicondylar and patellofemoral joint procedures annually.
replacements. Please note the difference in scale of
the y-axis between each plot. Figure 3.K1 (c) (page 141) shows that the volume of
unicondylar knee replacements has increased rapidly
Each bar in the figure is further stratified by the since 2014. Prior to 2020 the majority of additional
volume of procedures that the consultant conducted procedures were contributed by higher volume
in that year within that joint replacement type i.e. if a consultants i.e. those performing over 25 procedures
surgeon performed 25 elective total knee replacement annually. Only a very small proportion of consultants
procedures, 25 unicondylar knee replacements and contributed less than seven unicondylar knee
25 patellofemoral joint replacement procedures, replacements per year.
their annual total volume would be 75 procedures.
However, each 25 procedures are not aggregated Figure 3.K1 (d) shows that the volume of
and only contribute to the grey sub-division in each patellofemoral knee replacements has remained fairly
figure respectively. constant over the last ten years. Prior to 2020 the
majority of procedures recorded in the registry were
Figure 3.K1 (b) (page 140) shows that the volume contributed to by consultants who performed more
of total knee replacements has increased since than seven procedures annually.
Table 3.K2 Percentage of primary knee replacements by fixation, constraint, bearing and calendar year.
2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020
Fixation/bearing/ n= n= n= n= n= n= n= n= n= n= n= n= n= n= n= n= n=
constraint 41,585 42,517 50,345 67,032 74,468 76,654 79,247 82,826 86,720 86,440 96,242 100,081 105,164 107,136 103,144 106,572 50,904
All cemented 78.4 79.3 78.8 78.9 79.3 80.2 81.6 83.2 85.7 86.8 86.7 86.7 86.4 86.0 85.3 84.8 83.2
Cemented and
unconstrained, fixed 52.2 52.0 49.9 49.8 50.7 52.2 53.5 55.9 58.8 59.4 60.5 61.5 62.1 61.6 61.3 61.4 60.3
unconstrained, mobile 4.1 5.6 6.5 6.4 5.7 4.7 4.0 2.9 2.4 2.1 1.9 1.7 1.7 1.6 1.6 1.5 1.7
posterior-stabilised, fixed 20.3 19.3 19.6 19.8 20.4 20.8 21.2 21.1 20.8 20.9 20.2 19.9 19.3 19.5 19.1 18.6 17.9
posterior-stabilised,
1.0 1.6 1.8 1.6 1.4 1.4 1.4 1.2 1.1 1.2 1.0 0.8 0.6 0.4 0.3 0.3 0.4
mobile
constrained condylar 0.4 0.3 0.3 0.3 0.2 0.2 0.3 0.3 0.5 0.8 1.0 1.2 1.0 1.1 1.3 1.4 1.6
monobloc polyethylene
0.2 0.3 0.6 0.9 0.8 0.7 1.0 1.6 2.0 2.1 1.9 1.5 1.5 1.6 1.6 1.4 1.1
tibia
pre-assembled/hinged/
0.1 0.1 0.2 0.1 0.1 0.1 0.1 0.2 0.1 0.2 0.1 0.2 0.1 0.1 0.2 0.2 0.2
linked
All uncemented 6.3 5.9 6.3 6.3 6.0 5.5 4.6 4.0 3.2 2.5 2.5 2.3 2.0 2.0 1.8 1.9 1.8
Uncemented and
unconstrained, mobile 3.3 3.3 3.4 3.2 3.1 2.6 2.6 2.4 2.0 1.6 1.6 1.4 1.1 1.0 0.8 0.8 0.7
posterior-stabilised, fixed 0.6 0.5 0.5 0.4 0.3 0.3 0.2 0.2 0.2 0.2 0.3 0.2 0.1 0.2 0.2 0.1 0.1
other constraints <0.1 <0.1 <0.1 <0.1 <0.1 0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1
All hybrid 2.7 2.4 1.7 1.4 1.3 1.2 0.9 0.5 0.4 0.4 0.4 0.4 0.5 0.2 0.3 0.3 0.3
Hybrid and
unconstrained, mobile 0.3 0.2 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.2 0.2 0.3 0.3 0.1 0.1 0.1 <0.1
posterior-stabilised, fixed 0.1 0.1 0.1 0.1 0.1 0.1 0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 0.1 0.1 0.2
other constraints <0.1 0.2 0.2 0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 0
Note: Data from 2003 has been included in 2004, since 2003 was not a complete year.
143
144
Table 3.K2 (continued)
2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020
Fixation/bearing/ n= n= n= n= n= n= n= n= n= n= n= n= n= n= n= n= n=
constraint 41,585 42,517 50,345 67,032 74,468 76,654 79,247 82,826 86,720 86,440 96,242 100,081 105,164 107,136 103,144 106,572 50,904
All unicondylar,
8.0 8.2 8.8 8.3 8.3 8.0 7.8 7.1 6.9 6.6 6.4 6.0 5.8 6.0 6.7 7.1 7.9
cemented
Unicondylar, cemented and
fixed 0.8 1.0 1.0 1.0 1.2 1.4 1.8 1.9 2.3 2.7 3.0 3.3 3.6 4.1 5.0 5.7 6.7
mobile 6.5 6.2 6.6 6.4 6.4 6.0 5.5 4.7 4.1 3.4 3.0 2.5 1.9 1.7 1.4 1.2 1.1
0.7 0.9 1.2 0.9 0.7 0.6 0.5 0.4 0.5 0.4 0.4 0.3 0.3 0.3 0.3 0.2 0.2
tibia
All unicondylar,
0.1 0.2 0.2 0.3 0.4 0.7 0.9 1.2 1.2 1.4 2.0 2.8 3.4 3.9 4.3 4.3 5.1
uncemented/hybrid
Unicondylar, uncemented/hybrid and
fixed 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 <0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.2 0.2
mobile <0.1 0.1 0.1 0.2 0.3 0.5 0.7 1.0 1.1 1.4 1.9 2.7 3.3 3.8 4.1 4.1 4.8
<0.1 <0.1 <0.1 <0.1 <0.1 0.1 0.1 0.1 0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1
tibia
Patellofemoral 0.9 1.0 1.1 1.3 1.4 1.4 1.4 1.4 1.3 1.2 1.1 1.1 1.0 1.1 0.9 1.0 1.0
Multicompartmental <0.1 <0.1 <0.1 <0.1 <0.1 0.1 0.1 0.1 0.1 0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1
Unclassified 3.5 3.1 3.1 3.6 3.2 2.9 2.7 2.6 1.2 1.0 0.8 0.7 0.8 0.7 0.7 0.6 0.7
All 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100 100
Note: Data from 2003 has been included in 2004 since 2003 was not a complete year.
Table 3.K2 (page 143) shows the annual rates for knee replacements over time (now 2.1% of all knee
the usage of the different types of primary knee replacements). Usage of each implant of this type
replacements. Overall, more than 90% of all types has decreased proportionally to less than a quarter of
of primary knee replacement utilised all cemented those figures reported for 2004 (when they were 9.0%
fixation, and since 2004 the share of all implant of all knee replacements).
replacements of this type has increased by about
five percentage points. The main decline in the Figure 3.K2 illustrates the temporal changes in fixation,
type of primary knee replacements carried out has highlighting the dominance of cemented TKR primaries.
been in the use of all uncemented and hybrid total
Figure 3.K2 Fixation by year of procedure in primary knee replacement.
90
80
70

60
50
40
30
20
10
0
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020
Year of primary
Cemented Uncemented Hybrid Unicondylar Patellofemoral
Table 3.K3 Age at primary knee replacement by fixation, constraint and bearing type.
Age of patient (years) Percentage

Fixation Constraint and bearing type N Median (IQR)1 Mean (SD)2 male (%)3
All types 1,357,077 70 (63 to 76) 68.9 (9.6) 43.4
All cemented 1,136,212 70 (64 to 76) 69.7 (9.3) 42.4
Cemented and unconstrained, fixed 781,402 70 (64 to 76) 69.6 (9.1) 42.9
unconstrained, mobile 40,231 69 (62 to 76) 68.7 (9.6) 42.0
posterior-stabilised, fixed 270,635 70 (64 to 77) 69.8 (9.4) 41.1
posterior-stabilised, mobile 12,886 66 (60 to 74) 66.5 (10.1) 44.7
constrained condylar 10,698 71 (63 to 78) 69.9 (10.4) 36.2
monobloc polyethylene tibia 18,296 74 (69 to 79) 73.5 (8.2) 40.8
pre-assembled/hinged/linked 2,064 75 (66 to 82) 73.1 (12.6) 27.3
All uncemented 47,061 69 (62 to 75) 68.2 (9.6) 48.6
Uncemented and unconstrained, fixed 18,187 69 (61 to 75) 68.0 (9.8) 50.0
posterior-stabilised, fixed 3,428 67 (59 to 75) 66.7 (10.6) 53.0
other constraints 294 67 (60 to 73) 66.4 (9.0) 73.5
All hybrid 9,851 69 (62 to 76) 68.7 (9.8) 44.4

Hybrid and unconstrained, fixed 6,468 70 (63 to 76) 69.1 (9.5) 45.2
posterior-stabilised, fixed 819 68 (60 to 75) 67.0 (10.6) 46.4
other constraints 408 66 (58.5 to 75) 65.8 (10.7) 58.8
All unicondylar,
96,187 64 (57 to 71) 63.8 (9.8) 53.3
cemented
Unicondylar,
fixed 40,281 63 (56 to 70) 63.3 (10.0) 55.3
cemented and
mobile 49,610 64 (57 to 71) 64.2 (9.5) 51.6
monobloc polyethylene tibia 6,296 64 (57 to 71) 64.0 (10.1) 53.4
All unicondylar,
29,268 65 (58 to 72) 64.7 (9.6) 55.0
uncemented/hybrid
Unicondylar,
uncemented/hybrid fixed 1,259 66 (57 to 73) 65.2 (11.1) 43.8
and
mobile 27,606 65 (58 to 71) 64.7 (9.5) 55.7
monobloc polyethylene tibia 403 65 (59 to 72) 65.6 (9.1) 42.9
Patellofemoral 15,639 58 (50 to 67) 58.6 (11.7) 22.6
Multicompartmental 586 60 (53 to 67) 60.5 (10.1) 46.9
Unclassified 22,273 69 (61 to 75) 68.0 (10.3) 43.7
1
IQR = Interquartile range - age of middle 50% of patients at time of primary knee operation.
2
SD = Standard deviation.
3
The percentage male figures are based on a total number of primary knee replacements.
Table 3.K3 shows the age and gender distribution of uncemented and hybrid type procedures respectively.
patients undergoing primary knee replacement. The Conversely, cemented and uncemented unicondylar
median age of a person receiving a cemented TKR surgery was performed on a higher proportion of
was 70 years (IQR 64 to 76 years). Patients receiving males (53.3% and 55.0% respectively). Patellofemoral
cemented unicondylar prostheses were typically six surgery was predominantly carried out on females
years younger (median age 64 years; IQR 57 to 71) (77.4% of patients) who are typically younger than
compared to all types of knee replacement while those a TKR or unicondylar patient, with a median age at
receiving uncemented/hybrid unicondylar prostheses operation of 58.
were five years younger (median age 65 years; IQR
58 to 72). The patellofemoral group were typically 12 Table 3.K4 shows the ASA grade and indication for
years younger (median age 58 years; IQR 50 to 67) knee replacement by gender for all primary knee
compared to all types of knee replacement. Those replacements. ASA 2 is the most common ASA grade
receiving multicompartmental knee replacements were and only a small number of patients with a grade
typically ten years younger (median age 60 years; IQR greater than ASA 3 undergo knee replacement. The
53 to 67) compared to all types of knee replacement. majority of cases are performed with osteoarthritis as
the sole indication; 1,310,663 (96.6%) of all 1,357,077
Women were more likely to have a primary TKR; they knee replacements.
received 57.6%, 51.4% and 55.6% of cemented,
Table 3.K4 Primary knee replacement patient demographics.
Males N (%) Females N (%) All N (%)

Total 589,539 767,538 1,357,077
ASA 1 78,045 (13.2) 77,191 (10.1) 155,236 (11.4)
ASA 2 414,999 (70.4) 562,076 (73.2) 977,075 (72.0)

ASA 3 94,373 (16.0) 125,918 (16.4) 220,291 (16.2)
ASA 4 2,068 (0.4) 2,275 (0.3) 4,343 (0.3)
ASA 5 54 (<0.1) 78 (<0.1) 132 (<0.1)
Osteoarthritis as a
578,694 (98.2) 743,180 (96.8) 1,321,874 (97.4)
reason for primary
Osteoarthritis as
the sole reason for 573,703 (97.3) 736,960 (96.0) 1,310,663 (96.6)
primary
Mean (SD) Median (IQR) Mean (SD) Median (IQR) Mean (SD) Median (IQR)
Age
68.6 (9.3) 69 (62 to 75) 69.2 (9.8) 70 (63 to 76) 68.9 (9.6) 70 (63 to 76)
Note: Percentages in this table are calculated by column.
3.3.2 First revision after primary knee surgery
Figure 3.K3 (a) KM estimates of cumulative revision by year, in primary knee replacements.
6
5
2003
2004
4 2005
2006
2007
2008
3 2009
2010
2011
2012
2 2013
2014
2015
2016
1 2017
2018
2019
2020
0
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
Years since primary
In this section, estimates of cumulative revision in the of a knee joint being revised at three and five years
tables are presented at 1, 3, 5, 10, 15 and 17 years. increased for each operative year group between
A total of 40,451 first revisions of a knee prosthesis 2003 and 2008; the probability of being revised at
have been linked to registry primary knee replacement three and five years reduced for operations performed
surgery records of operations undertaken between between 2009 and 2020. From the peak in 2008, the
2003 and 2020. Figures 3.K3 (a) and (b) illustrate yearly survivorship curves are less divergent, i.e. a
temporal changes in the overall revision rates using slowing in the observed trend.
Kaplan-Meier estimates; procedures have been
grouped by the year of the primary operation. Possible reasons for a peak in the probability of
revision in the 2008 cohort are: 1) the registry was
Figure 3.K3 (a) plots each Kaplan-Meier survival curve not capturing the full range and number of operations
with a common origin, i.e. time zero is equal to the taking place in units in England and Wales until
year of operation. This illustrates that there was a small 2008, and 2) there could be bias in terms of the
increase in revision rates up until 2008, followed by a general overall health, risk of revision, and other key
small decline. characteristics of the patients on record in the registry
in the early years. Given that similar, more marked,
Figure 3.K3 (b) overleaf shows the same curves patterns are observed in primary hip replacements
plotted against calendar time, where the origin of and that the start of the reduction coincides with
each curve is the year of operation. It separates each the timeline of when NJR clinician feedback and
year enabling changes in failure rates to be clearly performance analyses were introduced, it is likely that
identified. In addition, the revision rates at 1, 3, 5, 7, these patterns represent improved survivorship as a
10, 13 and 15 years have been highlighted. If revision result of clinician feedback and the improved adoption
rates and timing of revision rates were static across of evidence-based practice.
time, it would be expected that all failure curves would
be the same shape and equally spaced; a departure
from this indicates a change in the number and timing
of revision procedures. The cumulative probability
150
Figure 3.K3 (b) KM estimates of cumulative revision by year, in primary knee replacements plotted by year of primary.
6

2
0
2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021
Year of primary
Cumulative probability of revision after primary:

1 year 3 years 5 years 7 years 10 years 13 years 15 years
Table 3.K5 KM estimates of cumulative revision (95% CI) by fixation, constraint and bearing, in primary knee replacements. Blue italics signify that
Fixation/constraint/ Time since primary

bearing type N 1 year 3 years 5 years 10 years 15 years 17 years
All types 1,357,077 0.49 (0.48-0.51) 1.75 (1.73-1.77) 2.54 (2.51-2.57) 4.13 (4.08-4.17) 5.84 (5.75-5.93) 6.43 (6.29-6.57)
Unclassified 22,273 0.73 (0.63-0.86) 2.23 (2.04-2.44) 3.16 (2.93-3.41) 5.35 (5.02-5.70) 7.15 (6.63-7.71) 8.23 (7.32-9.25)
All cemented 1,136,212 0.42 (0.41-0.43) 1.48 (1.45-1.50) 2.11 (2.08-2.14) 3.23 (3.19-3.27) 4.41 (4.33-4.50) 4.82 (4.69-4.96)
unconstrained, fixed 781,402 0.38 (0.37-0.39) 1.36 (1.34-1.39) 1.92 (1.89-1.96) 2.91 (2.86-2.96) 4.09 (3.99-4.19) 4.49 (4.33-4.64)
unconstrained, mobile 40,231 0.51 (0.44-0.58) 1.81 (1.68-1.95) 2.67 (2.51-2.84) 4.09 (3.87-4.31) 5.25 (4.92-5.61) 5.37 (5.00-5.76)
posterior-stabilised, fixed 270,635 0.48 (0.45-0.50) 1.69 (1.64-1.74) 2.50 (2.43-2.56) 3.89 (3.80-3.99) 5.09 (4.92-5.26) 5.63 (5.32-5.96)
posterior-stabilised, mobile 12,886 0.63 (0.51-0.78) 2.10 (1.87-2.37) 2.86 (2.57-3.17) 4.21 (3.84-4.62) 5.47 (4.86-6.16) 5.47 (4.86-6.16)
constrained condylar 10,698 0.90 (0.74-1.10) 2.09 (1.81-2.40) 2.71 (2.36-3.10) 4.02 (3.33-4.84) 5.39 (3.81-7.59) 6.49 (4.19-9.98)
monobloc polyethylene tibia 18,296 0.36 (0.28-0.46) 1.29 (1.13-1.48) 1.76 (1.56-1.98) 2.31 (2.05-2.61) 2.80 (2.37-3.31) 2.80 (2.37-3.31)
pre-assembled/hinged/linked 2,064 1.91 (1.39-2.63) 4.26 (3.41-5.32) 5.86 (4.79-7.16) 8.99 (7.17-11.25) 10.65 (8.21-13.75)
All uncemented 47,061 0.56 (0.50-0.64) 2.09 (1.96-2.23) 2.84 (2.68-3.00) 4.06 (3.86-4.27) 5.35 (5.05-5.68) 5.65 (5.23-6.10)
posterior-stabilised, fixed 3,428 0.62 (0.40-0.95) 2.33 (1.86-2.91) 3.30 (2.72-4.00) 5.71 (4.83-6.75) 7.90 (6.57-9.48) 8.81 (6.79-11.40)
other constraints 294 0.68 (0.17-2.71) 2.27 (1.02-5.00) 2.68 (1.28-5.55) 3.17 (1.59-6.26)
All hybrid 9,851 0.52 (0.39-0.68) 1.70 (1.46-1.99) 2.37 (2.08-2.71) 3.58 (3.19-4.02) 4.34 (3.85-4.90) 4.56 (3.93-5.28)
posterior-stabilised, fixed 819 0 2.14 (1.24-3.68) 3.67 (2.37-5.67) 5.77 (3.98-8.33) 6.23 (4.29-8.99)
other constraints 408 0.49 (0.12-1.96) 2.24 (1.17-4.26) 3.00 (1.71-5.22) 5.06 (3.19-7.98) 5.65 (3.57-8.88)
All unicondylar, cemented 96,187 0.98 (0.92-1.04) 3.71 (3.58-3.84) 5.62 (5.46-5.78) 10.41 (10.16-10.66) 15.74 (15.28-16.22) 17.59 (16.85-18.36)
fixed 40,281 0.63 (0.56-0.71) 2.62 (2.45-2.80) 4.04 (3.81-4.28) 7.55 (7.10-8.02) 11.38 (10.37-12.48) 12.17 (10.73-13.79)
mobile 49,610 1.28 (1.18-1.38) 4.37 (4.19-4.56) 6.49 (6.27-6.72) 11.67 (11.35-12.00) 17.30 (16.75-17.87) 19.30 (18.43-20.20)
monobloc polyethylene tibia 6,296 0.72 (0.54-0.97) 4.38 (3.89-4.94) 6.52 (5.90-7.19) 10.84 (9.98-11.78) 15.36 (14.03-16.80) 16.82 (14.94-18.91)
All unicondylar,
29,268 1.21 (1.09-1.35) 2.73 (2.53-2.94) 3.89 (3.63-4.18) 7.61 (6.95-8.32) 11.52 (9.88-13.42)
uncemented/hybrid
fixed 1,259 0.25 (0.08-0.78) 2.87 (1.99-4.14) 5.93 (4.46-7.86) 10.18 (7.89-13.10) 13.76 (10.43-18.04)
mobile 27,606 1.27 (1.14-1.41) 2.73 (2.53-2.95) 3.79 (3.52-4.07) 7.55 (6.82-8.35) 11.38 (9.41-13.74)
monobloc polyethylene tibia 403 0.50 (0.13-1.99) 2.29 (1.20-4.35) 3.36 (1.97-5.72) 6.73 (4.49-10.03)
Patellofemoral 15,639 1.05 (0.90-1.23) 5.75 (5.38-6.15) 9.42 (8.93-9.94) 18.23 (17.45-19.05) 25.36 (23.94-26.85) 26.71 (24.86-28.66)
Multicompartmental 586 1.04 (0.47-2.29) 7.19 (5.32-9.67) 9.84 (7.60-12.69) 14.02 (11.07-17.67)
151
Table 3.K5 on the previous page shows Kaplan-Meier
estimates of the cumulative percentage probability of
first revision, for any cause, for the cohort of all primary
knee replacements. This is broken down for TKR by
knee fixation type (cemented, uncemented or hybrid)
and sub-divided further within each fixation type by
the constraint (unconstrained, posterior-stabilised,
constrained condylar and highly constrained implants)
and bearing mobility (fixed or mobile) and for UKR,
by fixation type and bearing mobility (fixed or mobile).
The table shows updated estimates at 1, 3, 5, 10, 15
and 17 years from the primary operation together with
95% Confidence Intervals (95% CI).
Where groups have less than 250 cases remaining

at risk, the figures are shown in blue italics. Further
revisions in these groups would be highly unlikely,
and when they do occur, they may appear to have a
disproportionate impact on the Kaplan-Meier estimate,
i.e. the step upwards may seem steeper. Furthermore,
the upper 95% CI at these time points may be
underestimated. Although a number of statistical
methods have been proposed to deal with this, they
typically give different values and, as yet, there is no
clear consensus for the large datasets presented here.
Kaplan-Meier estimates are not shown at all when the
numbers at risk fell below ten.
Figures 3.K4 (a) to 3.K4 (d) illustrate the differences in with monobloc polyethylene tibias. The revision rates
revision rates between the types of knee replacement, in cemented TKRs that are posterior-stabilised and
fixation and constraint. It is worth noting the different those that have mobile bearings remain higher. The
vertical scales between the four figures. The results revision rates for UKRs remain substantially higher than
show the lowest revision rates for cemented for TKRs, this is most marked in the patellofemoral
unconstrained fixed bearing TKRs and cemented TKRs replacement and multicompartmental groups.
Figure 3.K4 (a) KM estimates of cumulative revision in primary total cemented knee replacements by
constraint and bearing. Blue italics in the numbers at risk table signify that fewer than 250 cases remained
at risk at these time points.
15
12

9
0
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
Years since primary
Unconstrained, fixed 781,402 663,507 513,082 371,029 253,363 157,388 85,943 35,810 9,248
Unconstrained, mobile 40,231 36,435 31,549 26,687 21,866 16,616 10,262 3,991 736
Posterior−stabilised, fixed 270,635 232,924 183,470 136,590 95,517 60,572 32,828 13,416 3,548
Posterior−stabilised, mobile 12,886 11,986 10,731 9,023 6,835 4,802 2,867 1,259 245
Constrained condylar 10,698 7,939 5,161 2,829 1,261 630 360 159 41
Monobloc polyethylene tibia 18,296 15,524 11,605 8,112 4,548 2,070 1,004 307 52
Pre−assembled/hinged/linked 2,064 1,528 1,092 715 408 238 123 48 12
Figure 3.K4 (b) KM estimates of cumulative revision in primary total uncemented knee replacements by
10
8
0
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
Years since primary
Unconstrained, fixed 18,187 16,001 13,623 11,543 9,811 7,467 4,492 1,715 450
Unconstrained, mobile 25,152 23,158 20,338 16,925 13,370 9,411 5,765 2,497 626
Posterior−stabilised, fixed 3,428 3,093 2,571 2,114 1,573 1,213 774 368 102
Figure 3.K4 (c) KM estimates of cumulative revision in primary total hybrid knee replacements by
10

0
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
Years since primary
Unconstrained, fixed 6,468 6,288 6,056 5,800 5,582 5,325 5,072 4,791 4,457 4,134 3,747 3,135 2,434 1,809 1,333 926 462 123
Unconstrained, mobile 2,156 2,100 1,989 1,860 1,679 1,355 1,068 858 692 564 456 377 284 201 146 92 56 14
Posterior−stabilised, fixed 819 726 562 487 451 420 396 376 342 307 263 213 160 121 80 45 18 7
Figure 3.K4 (d) KM estimates of cumulative revision in primary unicondylar or patellofemoral knee
replacements by fixation, constraint and bearing. Blue italics in the numbers at risk table signify that
30
25
20
15
10
0
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
Years since primary
Cemented, fixed 40,281 29,922 19,895 12,669 7,514 4,156 1,977 807 190
Cemented, mobile 49,610 45,876 40,751 34,697 27,704 19,975 11,956 5,082 1,365
Cemented, monobloc polyethylene tibia 6,296 5,716 4,891 4,076 3,250 2,408 1,596 755 151
Uncemented/hybrid, fixed 1,259 925 640 422 311 212 112 34 6
Uncemented/hybrid, mobile 27,606 20,124 11,598 5,581 2,721 1,146 340 89 13
Uncemented/hybrid, monobloc polyethylene tibia 403 383 367 323 270 145 39 5
Patellofemoral 15,639 13,490 10,769 8,196 5,987 3,808 2,078 776 193
Multicompartmental 586 517 445 371 288 152 43 8 4
Figure 3.K5 (a) KM estimates of cumulative revision in primary total knee replacements by gender
and age.
Males Females
15 15
12 12


9 9
6 6
3 3
0 0
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
<55y 29,403 25,403 20,369 15,368 11,037 7,239 4,168 1,849 505 41,825 36,232 28,632 21,534 15,312 10,024 5,827 2,610 730
55 to 59y 42,685 36,359 28,803 21,755 16,049 10,915 6,680 3,170 905 56,720 49,111 39,307 30,103 22,010 15,292 9,393 4,258 1,138
60 to 64y 77,356 66,980 54,113 42,161 31,644 21,458 12,480 5,467 1,458 92,339 80,926 65,691 51,410 38,366 25,894 15,222 6,476 1,720
65 to 69y 100,274 87,034 69,838 52,107 36,927 24,062 13,850 5,980 1,616 124,264 108,859 87,716 66,109 46,876 31,092 18,407 8,255 2,215
70 to 74y 106,552 90,093 69,664 51,159 35,853 23,011 12,532 4,890 1,160 139,632 119,569 93,569 70,477 50,788 33,610 19,136 8,021 2,059
75 to 79y 85,998 72,469 55,554 39,330 25,736 15,102 7,277 2,603 547 123,266 105,809 83,811 62,077 42,778 26,661 14,090 5,212 1,160
≥80y 66,376 53,313 37,580 23,490 13,034 6,191 2,281 628 73 106,435 89,188 66,884 45,504 28,023 14,600 6,155 1,830 318
Figure 3.K5 (a) shows that the chance of revision after

primary TKR is far higher in younger patient cohorts
and that men were slightly more likely, overall, to have
a first revision compared to women of comparable
grouped age, if they were under the age of 70 when
they underwent primary surgery.
Figure 3.K5 (b) KM estimates of cumulative revision in primary unicondylar knee replacements by
gender and age.
Males Females
30 30
25 25
20 20

15 15
10 10
5 5
0 0
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
<55y 10,410 8,674 6,670 4,902 3,468 2,232 1,184 477 116 11,711 9,797 7,483 5,498 3,880 2,557 1,419 596 130
55 to 59y 10,528 8,552 6,449 4,777 3,495 2,466 1,492 687 192 9,498 7,884 6,107 4,646 3,486 2,481 1,531 745 191
60 to 64y 13,329 11,058 8,626 6,653 5,018 3,533 2,098 870 233 10,095 8,518 6,785 5,274 4,022 2,774 1,652 660 170
65 to 69y 12,880 10,773 8,293 6,098 4,357 2,857 1,665 709 198 9,752 8,136 6,369 4,739 3,462 2,384 1,331 589 156
70 to 74y 10,132 8,091 5,804 4,153 2,927 1,931 1,028 404 90 8,302 6,708 4,898 3,562 2,609 1,799 1,051 468 128
75 to 79y 6,256 4,927 3,611 2,549 1,684 986 509 193 50 5,325 4,277 3,217 2,370 1,725 1,166 672 267 52
≥80y 3,820 2,849 1,887 1,169 731 383 159 38 11 3,417 2,702 1,943 1,378 906 493 229 69 8
Figure 3.K5 (b) shows that the risk of revision of primary Table 3.K6 shows gender and age stratified Kaplan-
unicondylar knee replacement is, again, substantially Meier estimates of the cumulative percentage
higher for younger patient cohorts but that there are probability of first revision, for any cause, firstly for all
less marked differences in younger patients in the risk cases combined, then by knee fixation / constraint /
of revision according to gender. The risk of revision is bearing sub-divisions. Estimates are shown, along with
higher in all age groups than it is for TKR; please note 95% CIs, for males and females within each of four
the differences in the vertical axes between Figures age bands, <55, 55 to 64, 65 to 74 and ≥75 years for
3.K5 (a) and (b). revision rate at 1, 3, 5, 10, 15 and 17 years after the
primary operation.
Table 3.K6 KM estimates of cumulative revision (95% CI) by gender, age, fixation, constraint and bearing, in primary knee replacements.
Males Females
Age at
Fixation/constraint/ primary
bearing type (years) N 1 year 3 years 5 years 10 years 15 years 17 years N 1 year 3 years 5 years 10 years 15 years 17 years
1.02 4.00 5.77 9.91 14.21 15.45 0.74 3.55 5.58 9.83 14.17 15.20
All cases <55 41,956 59,858
(0.93-1.13) (3.81-4.20) (5.53-6.02) (9.54-10.29) (13.54-14.90) (14.47-16.49) (0.67-0.81) (3.39-3.71) (5.38-5.79) (9.52-10.16) (13.58-14.78) (14.33-16.11)
1.64 5.15 7.57 13.63 16.53 18.43 1.54 4.95 8.20 12.81 18.16 18.16
Unclassified <55 927 1,246
(0.99-2.71) (3.87-6.84) (5.95-9.60) (11.27-16.44) (13.65-19.95) (14.16-23.79) (0.99-2.40) (3.86-6.34) (6.75-9.95) (10.88-15.05) (14.85-22.12) (14.85-22.12)
0.80 3.28 4.70 7.61 10.94 12.35 0.53 2.61 4.08 6.79 9.70 10.79
All cemented <55 27,144 39,381
(0.70-0.91) (3.06-3.51) (4.43-4.98) (7.21-8.03) (10.17-11.77) (11.09-13.74) (0.46-0.61) (2.45-2.78) (3.87-4.31) (6.47-7.14) (9.07-10.37) (9.71-11.98)
0.76 2.98 4.18 6.86 10.12 12.18 0.44 2.25 3.67 6.03 9.04 9.69
unconstrained, fixed <55 17,934 26,425
(0.64-0.90) (2.73-3.26) (3.87-4.51) (6.38-7.37) (9.14-11.19) (10.38-14.28) (0.37-0.53) (2.07-2.45) (3.42-3.93) (5.65-6.44) (8.25-9.89) (8.74-10.73)
1.04 4.14 6.08 8.60 11.88 11.88 0.75 2.96 4.93 7.58 10.11 10.95
unconstrained, mobile <55 1,362 1,747
(0.62-1.75) (3.18-5.37) (4.89-7.55) (7.11-10.39) (9.61-14.65) (9.61-14.65) (0.44-1.29) (2.24-3.89) (3.96-6.12) (6.29-9.11) (7.97-12.78) (8.40-14.20)
0.70 3.66 5.61 9.36 13.21 13.77 0.62 3.27 4.84 8.57 11.61 13.95
posterior-stabilised, fixed <55 6,562 9,553
(0.52-0.93) (3.21-4.17) (5.03-6.25) (8.47-10.33) (11.55-15.09) (11.84-15.98) (0.48-0.80) (2.91-3.66) (4.39-5.34) (7.84-9.36) (10.29-13.08) (10.84-17.85)
posterior-stabilised, 1.29 4.26 5.52 7.87 10.05 1.29 4.58 5.94 8.27 10.22
<55 706 786
mobile (0.67-2.46) (2.98-6.08) (4.03-7.54) (5.98-10.32) (7.14-14.06) (0.70-2.39) (3.31-6.32) (4.47-7.88) (6.43-10.61) (7.52-13.80)
2.09 4.63 5.71 7.60 0.42 2.26 2.64 4.51
constrained condylar <55 343 498
(1.00-4.33) (2.76-7.72) (3.49-9.28) (4.61-12.42) (0.10-1.65) (1.22-4.17) (1.45-4.77) (1.87-10.69)
monobloc polyethylene 0.61 4.52 4.52 6.66 1.11 3.57 5.24 5.24
<55 169 277
tibia (0.09-4.27) (2.18-9.25) (2.18-9.25) (3.44-12.68) (0.36-3.40) (1.87-6.76) (2.98-9.14) (2.98-9.14)
pre-assembled/hinged/ 2.94 4.59 8.75 13.46 4.28 9.24 12.28
<55 68 95
linked (0.74-11.25) (1.50-13.58) (3.68-20.01) (6.53-26.63) (1.63-11.00) (4.71-17.69) (6.73-21.83)
0.70 3.94 5.67 8.82 11.87 11.87 0.71 3.70 5.33 7.75 10.76 10.76
All uncemented <55 1,895 1,993
(0.40-1.20) (3.12-4.96) (4.66-6.90) (7.44-10.44) (9.85-14.27) (9.85-14.27) (0.42-1.20) (2.93-4.66) (4.37-6.48) (6.52-9.20) (8.88-13.00) (8.88-13.00)
0.87 4.21 5.74 8.48 11.86 1.01 3.09 4.16 6.84 9.76
unconstrained, fixed <55 816 811
(0.41-1.81) (2.98-5.95) (4.24-7.77) (6.49-11.03) (9.04-15.50) (0.51-2.01) (2.06-4.62) (2.90-5.95) (5.03-9.28) (6.97-13.58)
0.74 3.92 5.77 9.03 11.29 11.29 0.61 3.70 5.56 7.64 11.23
unconstrained, mobile <55 830 982
(0.33-1.63) (2.78-5.54) (4.32-7.68) (7.05-11.54) (8.45-15.01) (8.45-15.01) (0.28-1.36) (2.67-5.11) (4.25-7.26) (6.01-9.67) (8.63-14.55)
2.82 4.51 9.27 13.95 6.03 8.50 11.92 11.92
posterior-stabilised, fixed <55 224 0 194 0
(1.27-6.16) (2.36-8.52) (5.40-15.67) (8.32-22.88) (3.38-10.62) (5.20-13.72) (7.67-18.28) (7.67-18.28)
5.26 10.84
other constraints <55 25 6
(0.76-31.88) (2.82-36.85)
0.55 3.38 5.84 8.30 10.30 0.67 2.81 4.79 7.44 8.76 8.76
All hybrid <55 364 451
(0.14-2.19) (1.93-5.88) (3.81-8.92) (5.75-11.91) (7.16-14.71) (0.22-2.06) (1.60-4.89) (3.12-7.34) (5.22-10.57) (6.06-12.58) (6.06-12.58)
0.49 2.95 5.55 6.63 9.59 0.75 3.49 5.08 6.87 8.74
unconstrained, fixed <55 206 267
(0.07-3.40) (1.34-6.46) (3.11-9.80) (3.90-11.15) (5.82-15.57) (0.19-2.96) (1.83-6.60) (2.98-8.60) (4.32-10.84) (5.45-13.88)
2.82 4.62 11.94 0.98 2.00 3.20 7.62
unconstrained, mobile <55 71 103
(0.71-10.80) (1.50-13.76) (5.27-25.82) (0.14-6.76) (0.50-7.77) (1.04-9.63) (3.02-18.57)
2.86 9.61 2.44 7.32 9.76
posterior-stabilised, fixed <55 48 0 0 55 0
(0.41-18.60) (3.18-27.09) (0.35-16.08) (2.42-21.00) (3.78-23.94)
Note: Total sample on which results are based is 1,357,077 primary knee replacements.
159
160
Males Females
Age at
2.56 10.26 12.82 12.82 3.85 11.54
other constraints <55 39 26
(0.37-16.84) (3.98-25.06) (5.55-28.10) (5.55-28.10) (0.55-24.31) (3.87-31.64)
All unicondylar, 1.46 5.55 8.12 15.12 21.85 23.44 1.32 5.72 8.91 16.25 24.43 25.97
<55 8,229 9,284
cemented (1.22-1.75) (5.05-6.09) (7.51-8.79) (14.13-16.18) (20.14-23.69) (21.04-26.07) (1.11-1.58) (5.25-6.24) (8.30-9.57) (15.31-17.25) (22.64-26.34) (23.47-28.68)
1.06 3.86 5.72 9.79 13.34 0.75 4.01 6.19 13.26 17.77
fixed <55 4,091 4,187
(0.79-1.43) (3.27-4.55) (4.94-6.60) (8.44-11.35) (10.99-16.14) (0.52-1.07) (3.41-4.72) (5.38-7.11) (11.56-15.19) (14.89-21.14)
1.80 6.92 9.78 17.69 25.10 26.30 1.85 6.78 10.57 18.15 26.32 27.69
mobile <55 3,581 4,523
(1.41-2.29) (6.13-7.82) (8.83-10.83) (16.31-19.16) (22.91-27.48) (23.62-29.23) (1.49-2.28) (6.07-7.56) (9.68-11.53) (16.92-19.47) (24.20-28.60) (24.84-30.81)
monobloc polyethylene 2.18 7.72 12.06 21.11 26.28 1.23 8.38 11.64 17.57 29.83
<55 557 574
tibia (1.24-3.80) (5.74-10.34) (9.50-15.24) (17.43-25.45) (21.75-31.56) (0.59-2.56) (6.35-11.04) (9.19-14.70) (14.34-21.43) (23.35-37.63)
All unicondylar, 1.52 3.45 4.32 10.32 1.32 3.79 6.01 10.87
<55 2,181 2,427
uncemented/hybrid (1.08-2.15) (2.72-4.38) (3.44-5.41) (8.06-13.17) (0.93-1.88) (3.04-4.72) (4.93-7.32) (8.51-13.84)
3.49 6.81 14.77 5.38 7.10 14.47
fixed <55 103 0 131 0
(1.12-10.53) (2.83-15.91) (7.28-28.67) (2.26-12.50) (3.18-15.43) (6.53-30.37)
1.61 3.43 4.17 9.79 1.41 3.76 5.94 10.30
mobile <55 2,057 2,273
(1.14-2.28) (2.68-4.38) (3.29-5.29) (7.45-12.82) (0.99-2.00) (2.99-4.72) (4.83-7.30) (7.88-13.40)
monobloc polyethylene 5.00 5.00 4.35
<55 21 23
tibia (0.72-30.53) (0.72-30.53) (0.62-27.07)
2.58 9.71 14.04 23.29 34.94 0.84 6.24 10.01 20.30 27.90 28.61
Patellofemoral <55 1,151 4,970
(1.80-3.69) (8.06-11.67) (11.99-16.40) (20.29-26.65) (28.80-41.95) (0.62-1.14) (5.56-6.99) (9.12-10.97) (18.81-21.89) (25.33-30.68) (25.74-31.73)
8.00 9.64 9.64 0.96 9.74 15.04 21.15

Multicompartmental <55 65 0 106
(3.41-18.16) (4.45-20.21) (4.45-20.21) (0.14-6.63) (5.36-17.35) (9.34-23.72) (13.34-32.60)
0.71 2.48 3.60 5.73 8.22 9.28 0.47 2.17 3.27 5.42 7.66 8.45
All cases 55 to 64 147,781 175,504
(0.67-0.75) (2.39-2.56) (3.50-3.70) (5.58-5.89) (7.95-8.50) (8.82-9.77) (0.44-0.51) (2.10-2.25) (3.18-3.36) (5.28-5.55) (7.43-7.91) (8.10-8.81)
1.00 3.02 3.78 6.51 8.77 9.93 0.63 2.72 3.49 6.06 8.25 10.44
Unclassified 55 to 64 2,649 3,046
(0.68-1.46) (2.42-3.77) (3.09-4.61) (5.53-7.65) (7.44-10.33) (7.58-12.95) (0.40-0.99) (2.19-3.38) (2.87-4.23) (5.19-7.07) (6.95-9.79) (7.84-13.85)
0.62 2.21 3.18 4.77 6.58 7.26 0.39 1.83 2.71 4.20 5.82 6.29
All cemented 55 to 64 112,614 141,747
(0.58-0.67) (2.12-2.30) (3.07-3.29) (4.61-4.93) (6.31-6.86) (6.81-7.73) (0.35-0.42) (1.76-1.91) (2.62-2.80) (4.07-4.34) (5.59-6.07) (5.97-6.62)
0.55 2.05 2.91 4.27 6.09 6.71 0.36 1.69 2.43 3.73 5.41 5.97
unconstrained, fixed 55 to 64 78,517 98,177
(0.50-0.61) (1.94-2.15) (2.79-3.05) (4.09-4.46) (5.76-6.43) (6.24-7.21) (0.32-0.40) (1.61-1.78) (2.33-2.54) (3.58-3.89) (5.12-5.71) (5.57-6.40)
0.77 2.66 3.76 5.75 7.47 7.47 0.47 2.09 3.12 5.00 6.46 6.76
unconstrained, mobile 55 to 64 4,584 5,553
(0.55-1.07) (2.23-3.18) (3.23-4.38) (5.05-6.53) (6.40-8.72) (6.40-8.72) (0.32-0.69) (1.74-2.52) (2.67-3.63) (4.40-5.68) (5.60-7.45) (5.75-7.95)
0.76 2.59 3.82 5.94 8.00 9.05 0.44 2.19 3.40 5.26 6.76 7.05
posterior-stabilised, fixed 55 to 64 25,706 33,215
(0.66-0.88) (2.39-2.80) (3.57-4.08) (5.59-6.32) (7.42-8.63) (7.87-10.40) (0.37-0.51) (2.03-2.36) (3.20-3.62) (4.97-5.57) (6.28-7.26) (6.43-7.72)
posterior-stabilised, 0.86 2.47 3.06 4.81 5.46 5.46 0.33 1.79 3.03 4.63 6.65 6.65
55 to 64 1,882 2,141
mobile (0.53-1.40) (1.85-3.29) (2.35-3.96) (3.84-6.01) (4.25-6.99) (4.25-6.99) (0.16-0.69) (1.30-2.47) (2.36-3.87) (3.76-5.70) (5.26-8.39) (5.26-8.39)
Males Females
Age at
0.87 1.99 3.68 4.30 4.30 0.57 1.87 2.29 6.56 7.78
constrained condylar 55 to 64 936 1,287
(0.44-1.74) (1.21-3.24) (2.47-5.47) (2.76-6.67) (2.76-6.67) (0.27-1.19) (1.20-2.89) (1.52-3.46) (3.87-11.02) (4.54-13.14)
55 to 64 887 1,168
tibia (0.39-1.70) (1.21-3.19) (2.19-4.85) (2.79-6.20) (2.79-6.20) (0.08-0.80) (0.95-2.53) (1.54-3.57) (2.45-5.74) (3.49-8.85)
pre-assembled/hinged/ 5.07 12.01 14.95 21.35 2.52 3.11 4.44 8.45
55 to 64 102 206
linked (2.14-11.75) (6.81-20.74) (8.89-24.56) (12.44-35.23) (1.06-5.95) (1.41-6.80) (2.23-8.73) (4.35-16.07)
0.59 2.33 3.29 5.15 6.53 7.54 0.62 2.56 3.66 5.32 7.33 7.33
All uncemented 55 to 64 6,333 6,023
(0.43-0.81) (1.98-2.74) (2.86-3.79) (4.56-5.80) (5.76-7.40) (6.08-9.32) (0.45-0.86) (2.18-3.00) (3.19-4.19) (4.73-5.99) (6.40-8.38) (6.40-8.38)
0.52 2.52 3.44 5.70 7.14 7.14 0.68 2.90 3.73 5.47 7.26 7.26
(0.30-0.89) (1.96-3.24) (2.76-4.29) (4.74-6.85) (5.84-8.71) (5.84-8.71) (0.41-1.12) (2.26-3.71) (2.98-4.65) (4.52-6.62) (5.91-8.89) (5.91-8.89)
0.52 2.21 3.34 4.56 5.99 6.99 0.55 2.29 3.53 4.94 6.71 6.71
(0.32-0.85) (1.74-2.81) (2.74-4.07) (3.83-5.44) (5.00-7.17) (5.08-9.59) (0.34-0.86) (1.83-2.87) (2.93-4.25) (4.18-5.82) (5.48-8.21) (5.48-8.21)
1.15 2.01 2.39 5.97 7.11 0.99 3.06 4.47 7.97 12.48
posterior-stabilised, fixed 55 to 64 613 408
(0.55-2.40) (1.15-3.51) (1.42-4.00) (4.07-8.72) (4.82-10.43) (0.37-2.61) (1.75-5.33) (2.80-7.10) (5.50-11.48) (8.73-17.68)
1.82 3.95 3.95 6.48
other constraints 55 to 64 55 29 0 0 0
(0.26-12.21) (1.00-14.97) (1.00-14.97) (2.11-18.98)
0.37 1.62 3.06 4.71 7.05 8.36 0.55 2.26 3.23 5.10 5.47 5.47
All hybrid 55 to 64 1,093 1,289
(0.14-0.98) (1.01-2.60) (2.16-4.32) (3.51-6.30) (5.28-9.39) (5.66-12.27) (0.26-1.15) (1.57-3.26) (2.37-4.40) (3.94-6.60) (4.22-7.08) (4.22-7.08)
0.29 1.50 2.89 4.29 6.41 7.92 0.87 2.64 3.70 5.29 5.76 5.76
unconstrained, fixed 55 to 64 686 812
(0.07-1.17) (0.81-2.76) (1.86-4.50) (2.96-6.20) (4.49-9.12) (4.95-12.55) (0.42-1.81) (1.73-4.03) (2.59-5.28) (3.90-7.14) (4.26-7.76) (4.26-7.76)
0.45 0.45 2.03 2.03 3.81 1.29 1.64 5.99 5.99

unconstrained, mobile 55 to 64 221 324 0
(0.06-3.17) (0.06-3.17) (0.76-5.33) (0.76-5.33) (1.34-10.62) (0.48-3.39) (0.69-3.90) (2.82-12.49) (2.82-12.49)
2.46 4.03 7.67 2.50 5.33 8.43
posterior-stabilised, fixed 55 to 64 110 0 100 0
(0.61-9.59) (1.30-12.15) (3.20-17.75) (0.63-9.64) (2.03-13.61) (3.86-17.89)
1.32 5.28 6.62 12.75 1.92 1.92 1.92
other constraints 55 to 64 76 53
(0.19-8.97) (2.02-13.46) (2.81-15.17) (6.34-24.71) (0.27-12.88) (0.27-12.88) (0.27-12.88)
All unicondylar, 0.95 3.78 5.68 10.07 16.06 18.78 0.90 3.99 6.29 11.89 17.60 20.10
55 to 64 18,434 15,364
cemented (0.81-1.10) (3.50-4.08) (5.32-6.06) (9.53-10.64) (15.02-17.16) (16.91-20.83) (0.76-1.07) (3.68-4.33) (5.89-6.72) (11.27-12.55) (16.52-18.74) (18.46-21.87)
0.52 2.25 3.83 6.97 12.42 12.42 0.61 3.09 4.80 8.66 13.53
fixed 55 to 64 7,970 6,002
(0.38-0.71) (1.92-2.65) (3.35-4.39) (6.08-7.99) (9.98-15.41) (9.98-15.41) (0.44-0.85) (2.63-3.62) (4.18-5.50) (7.57-9.90) (10.91-16.71)
1.32 4.75 6.78 11.50 17.70 20.89 1.17 4.49 7.08 13.14 19.23 21.90
mobile 55 to 64 9,297 8,308
(1.11-1.58) (4.33-5.21) (6.27-7.32) (10.80-12.25) (16.45-19.03) (18.67-23.35) (0.96-1.43) (4.06-4.96) (6.53-7.67) (12.35-13.98) (17.95-20.59) (20.00-23.95)
55 to 64 1,167 1,054
tibia (0.35-1.38) (3.64-6.19) (5.53-8.63) (9.68-13.95) (13.62-19.43) (0.14-1.02) (3.45-6.08) (5.09-8.21) (9.82-14.29) (12.52-18.50)
161
162
Males Females
Age at
All unicondylar, 1.55 2.73 3.84 6.98 10.75 1.09 2.95 4.29 9.38 11.55
55 to 64 5,423 4,229
uncemented/hybrid (1.25-1.93) (2.30-3.24) (3.27-4.51) (5.69-8.54) (7.31-15.65) (0.81-1.47) (2.44-3.58) (3.59-5.11) (7.61-11.55) (8.96-14.84)
1.33 7.13 10.15 2.12 4.98 12.17
fixed 55 to 64 181 0 172 0
(0.33-5.25) (3.59-13.92) (5.43-18.57) (0.69-6.46) (2.24-10.90) (6.29-22.82)
1.63 2.81 3.74 6.72 10.19 1.14 2.98 4.25 9.30 9.78
mobile 55 to 64 5,189 3,969
(1.31-2.02) (2.37-3.34) (3.17-4.40) (5.36-8.41) (6.40-16.02) (0.85-1.53) (2.45-3.63) (3.53-5.10) (7.37-11.71) (7.69-12.41)
monobloc polyethylene 7.11 1.15 3.48 4.67 8.00
55 to 64 53 0 0 0 88
tibia (2.35-20.48) (0.16-7.88) (1.13-10.39) (1.78-11.96) (3.57-17.41)
1.82 5.83 10.66 22.38 28.95 0.85 5.41 9.43 17.63 24.70 27.44
Patellofemoral 55 to 64 1,124 3,701
(1.18-2.81) (4.56-7.45) (8.84-12.82) (19.30-25.86) (24.61-33.87) (0.60-1.21) (4.70-6.23) (8.46-10.51) (16.14-19.24) (22.24-27.38) (23.59-31.77)
6.45 8.35 12.69 1.94 8.14 10.54 18.01
Multicompartmental 55 to 64 111 0 105
(3.13-13.05) (4.43-15.43) (7.26-21.67) (0.49-7.54) (4.15-15.63) (5.79-18.76) (10.52-29.84)
0.53 1.65 2.30 3.57 4.90 5.24 0.36 1.44 2.12 3.36 4.41 4.74
All cases 65 to 74 234,380 288,646
(0.50-0.56) (1.60-1.71) (2.23-2.37) (3.47-3.67) (4.72-5.08) (4.98-5.52) (0.34-0.38) (1.40-1.49) (2.07-2.18) (3.28-3.45) (4.27-4.55) (4.54-4.95)
0.60 1.90 2.85 4.41 5.93 7.89 0.65 1.79 2.49 4.28 5.31 5.31
Unclassified 65 to 74 3,710 4,376
(0.40-0.91) (1.49-2.40) (2.34-3.46) (3.72-5.21) (4.81-7.30) (5.69-10.90) (0.45-0.94) (1.43-2.24) (2.06-3.02) (3.65-5.00) (4.42-6.38) (4.42-6.38)
0.47 1.49 2.08 3.09 4.17 4.40 0.32 1.27 1.86 2.79 3.51 3.76
All cemented 65 to 74 196,147 252,964
(0.44-0.50) (1.43-1.55) (2.01-2.15) (2.99-3.19) (3.99-4.36) (4.15-4.66) (0.29-0.34) (1.22-1.31) (1.80-1.92) (2.71-2.88) (3.38-3.65) (3.57-3.97)
0.45 1.38 1.90 2.75 3.81 4.02 0.27 1.16 1.68 2.53 3.29 3.51
(0.42-0.49) (1.31-1.44) (1.82-1.98) (2.64-2.87) (3.59-4.03) (3.71-4.35) (0.24-0.29) (1.11-1.22) (1.61-1.75) (2.44-2.63) (3.12-3.46) (3.26-3.77)
0.48 1.82 2.63 4.00 4.96 4.96 0.43 1.64 2.37 3.62 4.29 4.29
(0.34-0.68) (1.52-2.18) (2.26-3.07) (3.49-4.57) (4.24-5.80) (4.24-5.80) (0.31-0.59) (1.39-1.94) (2.06-2.74) (3.20-4.10) (3.74-4.91) (3.74-4.91)
0.53 1.72 2.43 3.83 4.99 5.28 0.40 1.42 2.20 3.29 3.93 4.32
posterior-stabilised, fixed 65 to 74 44,770 61,065
(0.46-0.60) (1.59-1.85) (2.28-2.59) (3.61-4.07) (4.61-5.41) (4.79-5.81) (0.35-0.45) (1.32-1.52) (2.07-2.33) (3.11-3.47) (3.69-4.19) (3.95-4.72)
posterior-stabilised, 0.45 1.91 2.59 3.37 4.05 0.58 1.90 2.57 3.79 4.85 4.85
65 to 74 2,025 2,431
mobile (0.23-0.86) (1.39-2.63) (1.97-3.42) (2.59-4.38) (3.08-5.31) (0.34-0.98) (1.43-2.54) (1.99-3.30) (3.01-4.77) (3.52-6.66) (3.52-6.66)
0.80 2.32 3.22 4.85 11.04 0.88 2.13 2.78 3.01 3.01
constrained condylar 65 to 74 1,442 2,455
(0.45-1.45) (1.60-3.35) (2.30-4.49) (3.16-7.42) (4.90-23.84) (0.57-1.34) (1.59-2.84) (2.11-3.66) (2.28-3.96) (2.28-3.96)
65 to 74 2,769 4,107
tibia (0.04-0.34) (1.08-2.06) (1.48-2.66) (1.92-3.37) (2.05-4.22) (0.21-0.59) (1.14-1.92) (1.59-2.54) (2.07-3.28) (2.17-3.70)
pre-assembled/hinged/ 3.12 8.22 10.34 13.83 1.23 3.44 5.23 6.60 6.60
65 to 74 165 342
linked (1.31-7.34) (4.74-14.06) (6.18-17.04) (8.67-21.70) (0.46-3.25) (1.86-6.34) (3.11-8.73) (3.71-11.61) (3.71-11.61)
0.57 1.81 2.31 3.35 4.25 4.25 0.48 2.24 2.97 3.82 4.54 5.07
All uncemented 65 to 74 8,955 8,926
(0.43-0.75) (1.55-2.12) (2.01-2.67) (2.95-3.82) (3.67-4.92) (3.67-4.92) (0.35-0.64) (1.95-2.58) (2.62-3.36) (3.40-4.28) (4.00-5.16) (4.20-6.12)
0.61 2.13 2.78 3.79 4.26 4.26 0.47 2.54 3.12 4.01 4.80 5.29
(0.40-0.94) (1.69-2.70) (2.25-3.42) (3.13-4.60) (3.49-5.19) (3.49-5.19) (0.29-0.78) (2.03-3.17) (2.54-3.82) (3.33-4.84) (3.93-5.85) (4.10-6.83)
Males Females
Age at
0.47 1.55 1.92 3.00 4.15 4.15 0.51 2.09 2.91 3.71 4.23 4.84
(0.31-0.71) (1.23-1.95) (1.55-2.37) (2.48-3.62) (3.33-5.17) (3.33-5.17) (0.35-0.74) (1.73-2.53) (2.47-3.43) (3.18-4.32) (3.59-4.97) (3.64-6.42)
0.99 2.05 2.89 3.81 5.28 0.20 2.13 2.86 3.94 6.05
posterior-stabilised, fixed 65 to 74 613 512
(0.45-2.19) (1.17-3.58) (1.77-4.69) (2.41-6.01) (3.15-8.81) (0.03-1.38) (1.15-3.93) (1.67-4.88) (2.39-6.46) (3.23-11.20)
1.09 2.32 2.32 2.32
other constraints 65 to 74 92 32 0 0 0
(0.15-7.47) (0.58-9.00) (0.58-9.00) (0.58-9.00)
0.41 1.65 1.93 2.83 3.18 3.18 0.55 1.67 1.90 2.69 3.28 3.28
All hybrid 65 to 74 1,724 2,006
(0.20-0.86) (1.13-2.40) (1.36-2.73) (2.08-3.85) (2.32-4.34) (2.32-4.34) (0.31-1.00) (1.18-2.35) (1.37-2.63) (2.02-3.58) (2.44-4.39) (2.44-4.39)
0.25 1.55 1.84 2.65 2.82 2.82 0.31 1.25 1.25 2.09 2.81 2.81
(0.08-0.78) (0.98-2.46) (1.20-2.80) (1.83-3.82) (1.95-4.07) (1.95-4.07) (0.11-0.81) (0.77-2.03) (0.77-2.03) (1.41-3.08) (1.91-4.14) (1.91-4.14)
1.28 1.97 2.35 3.53 5.54 1.47 3.23 4.06 4.87 4.87
unconstrained, mobile 65 to 74 317 479
(0.48-3.39) (0.89-4.33) (1.13-4.88) (1.80-6.84) (2.43-12.38) (0.70-3.05) (1.96-5.30) (2.57-6.39) (3.13-7.56) (3.13-7.56)
2.71 2.71 4.65 1.00 2.13 2.13 2.13
posterior-stabilised, fixed 65 to 74 132 0 162 0
(0.88-8.17) (0.88-8.17) (1.64-12.81) (0.14-6.89) (0.53-8.25) (0.53-8.25) (0.53-8.25)
other constraints 65 to 74 67 0 0 0 0 43 0 0
All unicondylar, 0.86 2.89 4.09 7.48 11.33 12.36 0.76 2.95 4.69 9.25 13.73 15.05
65 to 74 17,189 13,598
cemented (0.73-1.01) (2.63-3.16) (3.77-4.42) (6.97-8.02) (10.43-12.30) (10.85-14.06) (0.62-0.92) (2.67-3.27) (4.31-5.09) (8.63-9.91) (12.68-14.85) (13.60-16.65)
0.60 2.12 2.86 5.00 8.03 0.44 1.93 2.98 5.80 6.84
fixed 65 to 74 7,134 5,282

(0.45-0.82) (1.78-2.53) (2.42-3.37) (4.18-5.98) (6.15-10.45) (0.29-0.67) (1.56-2.40) (2.47-3.61) (4.73-7.12) (5.44-8.59)
1.10 3.35 4.66 8.52 12.50 13.81 1.03 3.53 5.56 10.75 15.74 16.96
mobile 65 to 74 8,952 7,472
(0.91-1.34) (2.99-3.75) (4.23-5.13) (7.87-9.22) (11.43-13.66) (11.94-15.94) (0.82-1.28) (3.13-3.98) (5.05-6.13) (9.97-11.59) (14.47-17.10) (15.34-18.74)
monobloc polyethylene 0.46 3.22 5.34 7.00 10.66 10.66 0.24 3.18 4.62 7.11 10.71
65 to 74 1,103 844
tibia (0.19-1.11) (2.30-4.50) (4.09-6.96) (5.48-8.92) (8.17-13.85) (8.17-13.85) (0.06-0.96) (2.16-4.67) (3.34-6.38) (5.36-9.39) (8.09-14.10)
All unicondylar, 1.13 2.38 3.27 5.46 0.97 2.75 4.18 8.25
65 to 74 5,823 4,456
0.60 5.37 8.39 11.57 2.13 6.26 8.91
fixed 65 to 74 176 216 0
(0.08-4.15) (2.71-10.48) (4.66-14.85) (6.59-19.87) (0.69-6.50) (2.99-12.87) (4.62-16.81)
1.14 2.24 3.01 5.23 1.03 2.78 4.03 9.11
mobile 65 to 74 5,583 4,154
(0.89-1.47) (1.85-2.70) (2.50-3.61) (4.08-6.71) (0.76-1.40) (2.27-3.39) (3.32-4.88) (6.76-12.22)
monobloc polyethylene 1.56 4.74 6.44 8.20 2.41 3.64 3.64
65 to 74 64 86 0
tibia (0.22-10.58) (1.55-13.97) (2.46-16.27) (3.49-18.63) (0.61-9.29) (1.19-10.88) (1.19-10.88)
163
164
Males Females
Age at
1.87 6.42 9.70 18.12 20.76 0.81 5.19 8.17 16.68 24.85 26.08
Patellofemoral 65 to 74 762 2,245
(1.11-3.14) (4.83-8.51) (7.64-12.28) (14.77-22.13) (15.32-27.81) (0.51-1.29) (4.31-6.23) (7.04-9.48) (14.83-18.73) (21.07-29.17) (21.76-31.06)
2.86 5.90 9.59 13.44 1.35 6.80 8.46 10.15
Multicompartmental 65 to 74 70 75
(0.72-10.94) (2.25-14.97) (4.39-20.24) (6.90-25.27) (0.19-9.21) (2.89-15.56) (3.88-17.93) (4.95-20.23)
0.43 1.12 1.47 2.14 2.64 2.69 0.39 1.02 1.40 2.04 2.53 2.66
All cases ≥75 165,422 243,530
(0.40-0.47) (1.07-1.18) (1.41-1.54) (2.04-2.24) (2.43-2.86) (2.46-2.95) (0.37-0.42) (0.98-1.07) (1.35-1.45) (1.97-2.12) (2.40-2.66) (2.48-2.85)
0.34 0.98 1.59 2.97 3.12 0.63 1.28 1.77 2.64 3.09 3.09
Unclassified ≥75 2,438 3,881
(0.17-0.67) (0.65-1.49) (1.13-2.23) (2.24-3.94) (2.35-4.14) (0.42-0.93) (0.96-1.70) (1.38-2.27) (2.11-3.30) (2.39-3.99) (2.39-3.99)
0.40 1.04 1.36 1.90 2.27 2.34 0.35 0.93 1.26 1.79 2.16 2.27
All cemented ≥75 145,501 220,714
(0.36-0.43) (0.99-1.10) (1.29-1.42) (1.81-2.00) (2.09-2.46) (2.12-2.58) (0.33-0.38) (0.89-0.97) (1.21-1.31) (1.72-1.86) (2.05-2.29) (2.11-2.44)
0.37 0.99 1.27 1.78 2.15 2.26 0.32 0.88 1.17 1.69 2.02 2.07
unconstrained, fixed ≥75 100,707 147,211
(0.33-0.41) (0.92-1.05) (1.19-1.35) (1.67-1.90) (1.93-2.39) (1.97-2.59) (0.29-0.35) (0.83-0.93) (1.11-1.23) (1.60-1.78) (1.89-2.16) (1.91-2.24)
0.38 1.00 1.51 1.96 2.18 0.42 0.94 1.37 1.86 2.21 2.21
unconstrained, mobile ≥75 4,367 7,509
(0.23-0.61) (0.74-1.37) (1.16-1.96) (1.52-2.52) (1.61-2.94) (0.30-0.60) (0.74-1.19) (1.11-1.68) (1.53-2.27) (1.77-2.76) (1.77-2.76)
0.47 1.17 1.56 2.23 2.66 2.66 0.38 1.02 1.42 1.99 2.50 2.77
posterior-stabilised, fixed ≥75 34,252 55,512
(0.40-0.55) (1.06-1.30) (1.42-1.71) (2.02-2.45) (2.30-3.08) (2.30-3.08) (0.33-0.43) (0.94-1.11) (1.32-1.53) (1.85-2.15) (2.22-2.81) (2.33-3.30)
posterior-stabilised, 0.53 1.39 1.52 1.84 1.84 0.52 1.01 1.37 1.99 2.61
≥75 1,147 1,768
mobile (0.24-1.18) (0.84-2.29) (0.93-2.47) (1.15-2.95) (1.15-2.95) (0.27-0.99) (0.63-1.62) (0.91-2.09) (1.33-2.98) (1.65-4.11)
0.82 1.79 2.01 2.59 1.13 1.76 1.94 2.27
constrained condylar ≥75 1,157 2,580
(0.43-1.56) (1.11-2.88) (1.25-3.25) (1.46-4.58) (0.78-1.63) (1.29-2.41) (1.42-2.65) (1.63-3.18)
monobloc polyethylene 0.28 1.02 1.25 1.57 0.45 0.83 1.10 1.43 1.43
≥75 3,643 5,276
tibia (0.15-0.52) (0.73-1.44) (0.91-1.72) (1.15-2.15) (0.30-0.67) (0.61-1.12) (0.83-1.46) (1.07-1.91) (1.07-1.91)
pre-assembled/hinged/ 0.46 2.54 3.26 4.43 1.39 2.82 3.79 6.04
≥75 228 858
linked (0.07-3.23) (1.06-6.00) (1.46-7.20) (2.03-9.52) (0.77-2.51) (1.82-4.35) (2.52-5.69) (3.69-9.81)
0.52 1.30 1.71 2.26 2.26 2.26 0.55 1.30 1.61 1.95 2.74 2.74
All uncemented ≥75 5,684 7,252
(0.36-0.75) (1.03-1.65) (1.38-2.11) (1.86-2.76) (1.86-2.76) (1.86-2.76) (0.40-0.75) (1.05-1.59) (1.33-1.95) (1.62-2.33) (2.17-3.47) (2.17-3.47)
0.61 1.02 1.47 1.83 1.83 0.77 1.59 2.01 2.01 2.71
unconstrained, fixed ≥75 2,227 2,791
(0.35-1.04) (0.67-1.56) (1.01-2.13) (1.28-2.60) (1.28-2.60) (0.50-1.18) (1.18-2.15) (1.53-2.64) (1.53-2.64) (1.94-3.79)
0.50 1.34 1.69 2.34 2.34 0.41 1.15 1.33 1.81 2.52 2.52
unconstrained, mobile ≥75 3,047 3,952
(0.30-0.83) (0.98-1.84) (1.27-2.25) (1.79-3.04) (1.79-3.04) (0.25-0.67) (0.85-1.54) (1.01-1.76) (1.40-2.35) (1.81-3.51) (1.81-3.51)
0.28 2.48 3.22 4.29 0.41 0.84 1.80 3.11 5.32
posterior-stabilised, fixed ≥75 366 498
(0.04-1.95) (1.24-4.90) (1.74-5.92) (2.42-7.55) (0.10-1.61) (0.32-2.23) (0.85-3.81) (1.63-5.89) (2.18-12.66)
2.44 2.44 2.44
other constraints ≥75 44 0 11 0
(0.35-16.08) (0.35-16.08) (0.35-16.08)
Males Females
Age at
0.43 0.89 1.28 2.17 2.17 0.65 1.27 1.50 1.99 1.99 1.99
All hybrid ≥75 1,189 1,735
(0.18-1.03) (0.48-1.66) (0.74-2.21) (1.34-3.52) (1.34-3.52) (0.36-1.16) (0.83-1.95) (1.00-2.23) (1.37-2.87) (1.37-2.87) (1.37-2.87)
0.38 0.91 1.42 2.35 2.35 0.62 1.18 1.50 2.01 2.01 2.01
unconstrained, fixed ≥75 825 1,142
(0.12-1.16) (0.43-1.89) (0.76-2.63) (1.37-4.01) (1.37-4.01) (0.30-1.30) (0.69-2.03) (0.92-2.44) (1.30-3.12) (1.30-3.12) (1.30-3.12)
0.93 0.93 0.93 0.93 0.96 1.23 1.23 1.83
unconstrained, mobile ≥75 216 425
(0.23-3.66) (0.23-3.66) (0.23-3.66) (0.23-3.66) (0.36-2.53) (0.51-2.93) (0.51-2.93) (0.77-4.33)
1.32 1.32 1.32 3.70 3.70 3.70
posterior-stabilised, fixed ≥75 90 0 122 0
(0.19-8.97) (0.19-8.97) (0.19-8.97) (1.19-11.22) (1.19-11.22) (1.19-11.22)
3.03
other constraints ≥75 58 0 0 0 46 0 0 0 0
(0.43-19.63)
All unicondylar, 0.81 2.12 2.98 5.17 7.91 7.91 1.10 2.96 4.44 7.60 10.16 11.24
≥75 7,404 6,685
cemented (0.63-1.05) (1.80-2.50) (2.57-3.46) (4.49-5.95) (6.02-10.37) (6.02-10.37) (0.87-1.38) (2.55-3.42) (3.92-5.02) (6.80-8.48) (8.70-11.85) (8.89-14.17)
0.51 1.26 2.00 3.17 5.77 0.72 2.33 3.43 5.21 5.21
fixed ≥75 3,094 2,521
(0.31-0.84) (0.89-1.79) (1.44-2.77) (2.10-4.76) (2.87-11.43) (0.45-1.16) (1.74-3.11) (2.62-4.47) (3.93-6.88) (3.93-6.88)
1.10 2.75 3.65 5.95 8.99 8.99 1.34 3.44 5.12 8.61 11.89 13.15
mobile ≥75 3,778 3,699
(0.81-1.49) (2.26-3.35) (3.07-4.34) (5.09-6.95) (6.73-11.95) (6.73-11.95) (1.02-1.77) (2.89-4.10) (4.43-5.92) (7.61-9.74) (10.04-14.06) (10.34-16.66)
≥75 532 465
tibia (0.10-1.51) (0.80-3.18) (1.49-4.60) (3.60-9.61) (3.60-9.61) (0.46-2.61) (1.06-3.87) (1.78-5.20) (3.57-8.62) (3.57-8.62)
All unicondylar, 1.34 2.10 2.53 4.46 0.70 1.90 2.83 4.59
≥75 2,672 2,057

1.19 1.19 1.19 0.55 2.12 3.24 4.82
fixed ≥75 91 189
(0.17-8.15) (0.17-8.15) (0.17-8.15) (0.08-3.82) (0.68-6.56) (1.19-8.66) (1.91-11.89)
1.36 2.17 2.64 4.65 0.73 1.93 2.88 5.04
mobile ≥75 2,546 1,835
(0.97-1.91) (1.64-2.88) (1.97-3.52) (2.78-7.72) (0.43-1.26) (1.35-2.76) (2.07-4.01) (3.20-7.91)
≥75 35 0 0 33 0 0
tibia
0.42 2.81 3.76 6.85 0.51 2.72 5.77 9.42 9.42
Patellofemoral ≥75 505 1,181
(0.11-1.67) (1.60-4.90) (2.27-6.20) (4.16-11.18) (0.23-1.14) (1.90-3.89) (4.45-7.46) (7.44-11.89) (7.44-11.89)
4.35 4.35
Multicompartmental ≥75 29 0 25 0 0 0
(0.62-27.07) (0.62-27.07)
165
Unicompartmental knee replacements seem to fare to undergo revision than their age-matched female
worse compared to TKR, with the chance of revision counterparts. The numbers for multicompartmental
at each estimated time point being approximately knee replacements are small in the age and gender
double or more than that of a TKR (Table 3.K5 on stratified groups but overall, the risk of revision is
page 151). The revision rate for cemented unicondylar markedly higher than that for total knee replacement
(medial or lateral UKR) knee replacements is 3.2 and more in keeping with patellofemoral replacement
times higher than the observed rate for cemented out to five years where the numbers at risk remain
TKR at ten years and 3.6 times higher at 17 years. above 250.
The revision rate for uncemented unicondylar (medial
or lateral UKR) knee replacements is 2.4 times 3.3.3 Revisions after primary knee
higher than for cemented TKR at ten years and 2.6 replacement surgery by main brands
times higher at 15 years, although the numbers
for TKR and UKR
for the last estimate are small and so we suggest
should be treated with caution. The revision rate for As in previous reports, only brands that have been
patellofemoral replacement is 5.6 times higher than for used in a primary knee replacement in 1,000 or more
cemented TKR at ten years and 5.5 times higher at 17 operations have been included (Tables 3.K7 (a) and (b)
years although again, we advise a degree of caution and Table 3.K8 (on page 174)). Table 3.K7 (b) shows
since the number of patellofemoral replacements at a breakdown of these included brands according to
risk at 17 years is small. Multicompartmental knee whether the patella was resurfaced or not at the time
replacements have relatively small numbers, and at of the primary procedure. In Table 3.K9 (a) (page 175)
five years the risk of revision is 4.5 times higher than brands are displayed with a breakdown according to
for cemented TKR, 1.8 times higher than for cemented fixation, constraint and bearing mobility where there
unicondylar knee replacements and 2.5 times higher are more than 2,500 operations for TKR and more
than for uncemented unicondylar knee replacements. than 1,000 operations for UKR. Table 3.K9 (b) (page
The rates are approximately equivalent to those seen 179) provides an additional breakdown for the TKRs
for patellofemoral replacements. displayed in Table 3.K9 (a) according to whether the
patella was resurfaced or not. Further breakdowns
First revision of an implant is slightly less likely in
by component are available from other sources of
females than in males overall for the most commonly
information, such as ODEP. The figures in blue italics
used fixation method (cemented) but, broadly, a
are at time points where fewer than 250 primary knee
patient from a younger age group is more likely to
replacements remain at risk. No results are shown
be revised irrespective of gender, with the youngest
where the number had fallen below ten cases. We
group having the worst predicted outcome in terms
have made no attempt to adjust for other factors that
of the risk of subsequent revision (Table 3.K6 on page
may influence the chance of revision, so the figures are
159). Conversely, female patients are more likely to
unadjusted probabilities. Given that the sub-groups
have a unicondylar implant revised in the longer term
may differ in composition with respect to age and
compared to their male, age-equivalent counterparts,
gender, the percentage of males and the median (IQR)
except for when under the age of 55. For
of the ages are also shown in these tables.
patellofemoral implants, males are generally more likely
Table 3.K7 (a) KM estimates of cumulative revision (95% CI) by total knee replacement brands. Blue italics signify that
Median Time since primary
(IQR)
age at Percentage
Brand1 N primary (%) male 1 year 3 years 5 years 10 years 15 years 17 years
All total knee 70 0.43 1.50 2.15 3.27 4.45 4.85
replacements
1,193,125 43
(63 to 76) (0.41-0.44) (1.48-1.53) (2.12-2.18) (3.23-3.31) (4.37-4.53) (4.72-4.97)
68 0.78 2.70 3.29 4.72
ACS PC[Fem]ACS[Tib] 1,159 50
(61 to 73) (0.41-1.50) (1.89-3.84) (2.38-4.54) (3.48-6.40)
Advance MP
69 0.07 1.67 2.53 3.38
Stature[Fem] 1,502 13
(62 to 75) (0.01-0.47) (1.12-2.48) (1.82-3.51) (2.40-4.76)
Advance[Tib]
Advance MP[Fem] 70 0.58 2.08 2.95 4.24 4.99 6.47
8,926 48
Advance[Tib] (64 to 76) (0.44-0.76) (1.80-2.40) (2.60-3.34) (3.78-4.76) (4.33-5.73) (4.08-10.19)
Advance PS[Fem] 72 0.56 2.64 3.36 5.97 7.24
1,438 45
Advance[Tib] (66 to 77) (0.28-1.12) (1.90-3.67) (2.50-4.51) (4.57-7.77) (5.48-9.53)
71 0.31 1.52 2.19 3.49 5.62 6.43
AGC V2[Fem:Tib] 39,003 43
(65 to 77) (0.26-0.37) (1.40-1.65) (2.05-2.35) (3.29-3.69) (5.25-6.01) (5.85-7.06)
71 0.30 1.59 2.23 3.57 5.83 6.98
AGC[Fem]AGC V2[Tib] 28,816 42
(64 to 77) (0.24-0.37) (1.45-1.74) (2.06-2.41) (3.33-3.83) (5.28-6.44) (5.63-8.64)
AS Columbus
65 0.41 1.50 2.42
Cemented[Fem] 1,260 52
(59 to 71.5) (0.17-0.99) (0.88-2.54) (1.46-3.98)
Columbus CR/PS[Tib]
Attune[Fem] 69 0.40 1.50 2.10
28,721 43

Attune FB[Tib] (62 to 76) (0.33-0.48) (1.34-1.67) (1.88-2.35)
Attune[Fem] 69 0.19 0.90 1.34
4,953 44
Attune RP[Tib] (62 to 76) (0.10-0.37) (0.64-1.27) (0.96-1.88)
Columbus
70 0.43 1.48 2.05 3.00 3.71
(64 to 76) (0.34-0.55) (1.29-1.69) (1.82-2.32) (2.63-3.43) (3.06-4.50)
Columbus CR/PS[Tib]
E-Motion Bicondylar
68 0.66 2.35 3.33 4.56 5.45
Knee[Fem] 3,339 45
(61 to 74) (0.44-1.00) (1.88-2.93) (2.75-4.02) (3.82-5.45) (4.48-6.61)
E-Motion FP[Tib]
Endo-Model
76 1.27 3.30 5.05 7.95 9.91
Standard Rotating 1,338 28
(68 to 83) (0.78-2.07) (2.41-4.51) (3.86-6.58) (5.98-10.54) (7.19-13.58)
Hinge[Fem:Tib]
70 0.46 1.64 2.11
EvolutionMP[Fem:Tib] 1,815 45
(63 to 76) (0.23-0.93) (1.07-2.49) (1.39-3.20)
Genesis II
59 0.58 2.43 3.57 6.06 7.67 7.67
Oxinium[Fem] 11,362 40
(54 to 65) (0.46-0.74) (2.15-2.75) (3.21-3.97) (5.50-6.67) (6.78-8.67) (6.78-8.67)
Genesis II[Tib]
71 0.46 1.50 2.05 3.02 3.49 3.83
Genesis II[Fem:Tib] 85,534 42
(65 to 77) (0.42-0.51) (1.42-1.59) (1.95-2.16) (2.86-3.18) (3.21-3.79) (3.30-4.44)
Insall-Burstein II
Microport[Fem] 71 0.35 1.74 2.93 5.14 7.07 7.74
2,020 45
Insall-Burstein (65 to 77) (0.17-0.73) (1.25-2.43) (2.26-3.79) (4.21-6.28) (5.88-8.49) (6.38-9.38)
(Microport)[Tib]
Journey II BCS
66 0.58 2.42 2.60
(59 to 73) (0.38-0.87) (1.90-3.08) (2.04-3.31)
Journey[Tib]
71 0.25 1.74 2.70 4.73 6.68 7.20
Kinemax[Fem:Tib] 10,915 43
(64 to 77) (0.17-0.36) (1.51-2.01) (2.40-3.03) (4.32-5.17) (6.15-7.25) (6.60-7.85)
LCS Complete[Fem] 70 0.43 1.69 2.49 3.64 4.22
29,139 44
M.B.T.[Tib] (63 to 76) (0.36-0.51) (1.54-1.84) (2.31-2.69) (3.40-3.90) (3.90-4.57)
70 0.65 1.78 2.32 3.00 3.80 4.08
LCS[Fem:Tib] 2,001 41
(63 to 76) (0.38-1.12) (1.28-2.48) (1.74-3.09) (2.31-3.88) (2.98-4.83) (3.19-5.20)
Brands shown have been used in at least 1,000 primary total knee replacement operations.
1
Note: Blank cells indicate the number at risk is below ten and therefore estimates are omitted as they are unreliable.
Note: Femoral brand precedes [Fem], tibial brand precedes [Tib]. [Fem:Tib] indicates the same brand for both femoral and tibial component.
Table 3.K7 (a) (continued)
(IQR)
age at Percentage
Legion CR COCR[Fem] 71 0.48 1.61 2.13
1,045 44
Genesis II[Tib] (65 to 77) (0.20-1.15) (0.99-2.62) (1.37-3.29)
70 0.41 1.77 2.77 5.47 9.03 12.39
Maxim[Fem:Tib] 1,744 43
(63 to 77) (0.19-0.85) (1.24-2.53) (2.08-3.68) (4.41-6.77) (7.33-11.10) (8.95-17.03)
70 0.32 1.17 1.63 2.70 3.23 3.23
MRK[Fem:Tib] 15,118 44
(64 to 77) (0.24-0.42) (1.00-1.37) (1.42-1.86) (2.36-3.09) (2.69-3.87) (2.69-3.87)
Natural Knee II[Fem] 70 0.32 1.34 2.22 4.02 6.85 7.25
2,814 42
NK2[Tib] (64 to 76) (0.17-0.62) (0.97-1.85) (1.72-2.84) (3.31-4.88) (5.55-8.44) (5.79-9.08)
Nexgen LCCK[Fem] 71 1.13 2.65 3.33 4.93 9.65
1,091 36
Nexgen[Tib] (64 to 79) (0.65-1.99) (1.79-3.91) (2.30-4.81) (3.23-7.48) (4.28-20.94)
70 0.38 1.29 2.03 3.44 4.55 4.93
Nexgen[Fem:Tib] 174,049 42
(64 to 76) (0.35-0.41) (1.24-1.35) (1.96-2.11) (3.33-3.56) (4.34-4.78) (4.61-5.28)
Nexgen[Fem]
LPS (Legacy 67 0.47 1.90 2.60 4.39 6.24 6.24
3,239 46
Posterior Stabilised (59 to 74) (0.28-0.77) (1.47-2.44) (2.09-3.24) (3.65-5.29) (5.02-7.74) (5.02-7.74)
ZimmerBiomet)[Tib]
Nexgen[Fem] 64 0.61 2.64 3.33 4.38 4.95 4.95
4,273 57
TM Monoblock[Tib] (58 to 71) (0.42-0.90) (2.19-3.18) (2.82-3.93) (3.77-5.08) (4.23-5.78) (4.23-5.78)
Optetrak CR[Fem] 70 0.86 3.45 4.90 8.06 9.29

1,639 43
Optetrak[Tib] (63 to 76) (0.51-1.45) (2.66-4.46) (3.94-6.09) (6.71-9.67) (7.23-11.89)
Persona CR[Fem] 70 0.25 0.74 1.68
4,722 46
Persona[Tib] (63 to 76) (0.13-0.46) (0.43-1.26) (0.95-2.94)
Persona PS[Fem] 70 0.44 1.63 3.11
1,446 42
Persona[Tib] (63 to 76) (0.20-0.97) (1.02-2.60) (2.05-4.72)
PFC Sigma Bicondylar
65 0.63 2.00 2.78 3.96 4.97 5.15
Knee[Fem] 17,259 47
(58 to 72) (0.52-0.76) (1.80-2.23) (2.54-3.04) (3.65-4.29) (4.50-5.49) (4.58-5.79)
M.B.T.[Tib]
70 0.39 1.29 1.78 2.51 3.28 3.55
Knee[Fem] 170,400 43
(64 to 76) (0.36-0.42) (1.23-1.35) (1.71-1.85) (2.43-2.60) (3.16-3.42) (3.36-3.76)
PFC Bicondylar[Tib]
Knee[Fem] 70 0.37 1.41 1.96 2.66 2.89
192,189 42
PFC Sigma (64 to 77) (0.34-0.40) (1.36-1.47) (1.90-2.03) (2.56-2.75) (2.73-3.07)
Bicondylar[Tib]
73 0.41 1.37 1.87 2.72 3.76 4.11
Profix[Fem:Tib] 3,956 44
(67 to 78) (0.25-0.67) (1.05-1.79) (1.48-2.35) (2.23-3.31) (2.95-4.78) (3.12-5.40)
70 0.65 3.03 3.90 6.49 8.54 8.54
Rotaglide +[Fem:Tib] 1,999 44
(63 to 76) (0.38-1.13) (2.36-3.90) (3.12-4.86) (5.43-7.74) (7.21-10.10) (7.21-10.10)
71 0.56 2.41 3.79 4.34 6.25
Rotaglide[Fem:Tib] 1,449 39
(63 to 77) (0.28-1.11) (1.73-3.35) (2.89-4.96) (3.34-5.62) (4.60-8.47)
69 0.55 1.29 1.42
Saiph[Fem:Tib] 1,855 39
(63 to 75) (0.29-1.05) (0.81-2.06) (0.89-2.25)
70 0.41 1.59 2.43 3.69
Scorpio NRG[Fem:Tib] 14,101 43
(64 to 76) (0.32-0.53) (1.40-1.82) (2.18-2.70) (3.35-4.06)
68 0.37 2.17 3.12 4.69 6.06 6.06
Scorpio[Fem:Tib] 3,255 45
(61 to 75) (0.21-0.65) (1.72-2.74) (2.57-3.80) (3.98-5.51) (5.03-7.28) (5.03-7.28)
Scorpio[Fem] 71 0.44 1.83 2.63 4.02 5.15 5.34
21,689 42
Scorpio NRG[Tib] (64 to 77) (0.36-0.54) (1.66-2.02) (2.42-2.85) (3.76-4.31) (4.81-5.52) (4.94-5.77)
Sphere[Fem] 69 0.81 2.10 2.98
1,698 44
GMK[Tib] (62 to 75) (0.47-1.39) (1.46-3.03) (2.10-4.22)
1

(IQR)
age at Percentage
70 0.68 1.79 2.38 3.52 4.80 5.21
TC Plus[Fem:Tib] 16,030 45
(64 to 76) (0.57-0.82) (1.60-2.01) (2.15-2.63) (3.23-3.84) (4.35-5.30) (4.58-5.93)
70 0.49 1.47 2.06 3.04 4.38
Triathlon[Fem:Tib] 145,056 43
(63 to 76) (0.46-0.53) (1.40-1.54) (1.97-2.15) (2.90-3.19) (3.55-5.40)
Unity Knee[Fem] 70 0.28 0.80 1.06
1,458 45
Unity[Tib] (63 to 76) (0.11-0.75) (0.43-1.50) (0.59-1.88)
70 0.39 1.44 2.04 2.98
Vanguard[Fem:Tib] 82,502 42
(63 to 76) (0.35-0.43) (1.36-1.53) (1.94-2.16) (2.78-3.20)
1
Table 3.K7 (b) KM estimates of cumulative revision (95% CI) in total knee replacement brands by whether a patella
component was recorded. Blue italics signify that fewer than 250 cases remained at risk at these time points.
(IQR)
Patella age at Male
Brand1 status N primary (%) 1 year 3 years 5 years 10 years 15 years 17 years
With 70 0.41 1.27 1.84 2.88 4.00 4.28
458,640 38
All total knee Patella (63 to 76) (0.39-0.43) (1.24-1.31) (1.80-1.88) (2.82-2.95) (3.87-4.13) (4.10-4.47)
replacements Without 70 0.43 1.64 2.33 3.49 4.71 5.18
734,485 45
Patella (63 to 76) (0.42-0.45) (1.61-1.67) (2.29-2.37) (3.44-3.55) (4.61-4.81) (5.01-5.34)
ACS PC[Fem] With 68 2.25 3.57 3.57
90 28
ACS[Tib] Patella (61 to 74) (0.57-8.69) (1.16-10.70) (1.16-10.70)
Without 68 0.66 2.61 3.24 4.72
1,069 52
Patella (61 to 73) (0.31-1.38) (1.80-3.79) (2.32-4.54) (3.44-6.46)

Advance MP
With 69 0.62 1.72 2.07
Stature[Fem] 508 12 0
Patella (62 to 75) (0.20-1.93) (0.82-3.60) (1.03-4.14)
Advance[Tib]
Without 69 0.10 2.19 2.96 4.00
994 14
Patella (62 to 75) (0.01-0.72) (1.43-3.34) (2.05-4.26) (2.72-5.86)
Advance MP[Fem] With 70 0.53 1.52 2.10 3.34 3.95
3,048 43
Advance[Tib] Patella (63 to 76) (0.32-0.86) (1.14-2.03) (1.63-2.71) (2.66-4.18) (3.02-5.17)
Without 70 0.60 2.37 3.39 4.68 5.52 8.67
5,878 50
Patella (64 to 76) (0.43-0.84) (2.00-2.80) (2.94-3.91) (4.09-5.36) (4.71-6.46) (4.23-17.31)
Advance PS[Fem] With 71 0.80 3.99 5.00 8.78 8.78
252 36
Advance[Tib] Patella (66 to 76) (0.20-3.17) (2.09-7.54) (2.79-8.86) (5.32-14.31) (5.32-14.31)
Without 72 0.51 2.36 3.02 5.37 6.92
1,186 48
Patella (66 to 78) (0.23-1.13) (1.61-3.45) (2.14-4.25) (3.92-7.33) (4.99-9.55)
With 71 0.25 1.24 1.84 3.01 4.68 4.85
AGC V2[Fem:Tib] 12,157 35
Patella (65 to 77) (0.17-0.36) (1.06-1.46) (1.61-2.10) (2.70-3.37) (4.08-5.35) (4.18-5.61)
Without 71 0.34 1.65 2.35 3.69 5.99 7.02
26,846 46
Patella (65 to 77) (0.28-0.42) (1.50-1.81) (2.17-2.55) (3.46-3.95) (5.55-6.47) (6.29-7.83)
AGC[Fem] With 71 0.25 1.19 1.69 2.98 6.00 6.00
9,725 37
AGC V2[Tib] Patella (64 to 77) (0.17-0.37) (0.99-1.43) (1.44-1.98) (2.61-3.40) (5.04-7.14) (5.04-7.14)
Without 71 0.33 1.79 2.50 3.87 5.53 7.57
19,091 45
Patella (64 to 77) (0.26-0.42) (1.61-1.99) (2.28-2.74) (3.57-4.20) (4.94-6.19) (5.43-10.50)
1
Table 3.K7 (b) (continued)

(IQR)
Patella age at Male
AS Columbus
With 65 0.13 1.20 1.96
Cemented[Fem] 792 51
Patella (59 to 71) (0.02-0.95) (0.57-2.53) (0.97-3.93)
Columbus CR/PS[Tib]
Without 64 0.89 2.04 3.32
468 52
Patella (58 to 72) (0.34-2.37) (0.95-4.36) (1.61-6.76)
Attune[Fem] With 70 0.35 1.24 1.86
13,599 39
Attune FB[Tib] Patella (63 to 76) (0.26-0.47) (1.04-1.47) (1.54-2.24)
Without 69 0.45 1.72 2.32
15,122 47
Patella (62 to 76) (0.35-0.57) (1.50-1.98) (2.02-2.67)
Attune[Fem] With 69 0.20 0.80 1.08
3,190 40
Attune RP[Tib] Patella (62 to 76) (0.09-0.43) (0.51-1.26) (0.68-1.70)
Without 69 0.19 1.07 1.78
1,763 52
Patella (62 to 76) (0.06-0.59) (0.63-1.81) (1.10-2.88)
Columbus
With 70 0.61 1.32 1.65 3.19 5.85
Patella (64 to 76) (0.42-0.88) (1.02-1.71) (1.29-2.09) (2.25-4.50) (3.48-9.76)
Columbus CR/PS[Tib]
Without 71 0.36 1.54 2.21 3.03 3.32
11,214 45
Patella (65 to 77) (0.26-0.49) (1.32-1.80) (1.93-2.54) (2.63-3.50) (2.81-3.92)
E-Motion Bicondylar 7.97 7.97
With 66 1.05 5.63
Knee[Fem] 289 33 (5.31- (5.31-
Patella (60 to 73) (0.34-3.21) (3.49-9.03)
E-Motion FP[Tib] 11.87) 11.87)
Without 68 0.63 2.03 2.87 4.16 5.06

3,050 46
Patella (61 to 74) (0.40-0.98) (1.58-2.61) (2.32-3.56) (3.42-5.05) (4.09-6.24)
Endo-Model
With 75 1.55 3.04 4.69 7.34
Standard Rotating 271 28
Patella (66 to 82) (0.58-4.07) (1.45-6.31) (2.53-8.61) (3.79-13.98)
Hinge[Fem:Tib]
Without 76 1.20 3.37 5.14 8.11 9.49
1,067 28
Patella (69 to 83) (0.69-2.11) (2.38-4.76) (3.82-6.90) (5.91-11.09) (6.79-13.19)
EvolutionMP With 71 0.63 1.55 1.55
686 46
[Fem:Tib] Patella (65 to 77) (0.24-1.68) (0.71-3.36) (0.71-3.36)
Without 68 0.36 1.64 2.25
1,129 45
Patella (62 to 75) (0.14-0.97) (0.98-2.71) (1.39-3.63)
Genesis II
With 59 0.50 1.77 2.37 4.21 5.51
Patella (54 to 65) (0.35-0.71) (1.45-2.16) (1.98-2.83) (3.57-4.95) (4.35-6.97)
Genesis II[Tib]
Without 59 0.68 3.18 4.90 8.03 9.90
5,232 44
Patella (54 to 65) (0.49-0.94) (2.72-3.72) (4.30-5.58) (7.14-9.03) (8.60-11.37)
With 71 0.46 1.25 1.63 2.38 2.78 3.26
Genesis II[Fem:Tib] 39,672 38
Patella (65 to 77) (0.40-0.53) (1.14-1.37) (1.50-1.78) (2.18-2.60) (2.44-3.16) (2.39-4.44)
Without 71 0.47 1.71 2.39 3.50 4.01 4.22
45,862 46
Patella (65 to 77) (0.41-0.54) (1.59-1.84) (2.24-2.55) (3.28-3.74) (3.62-4.45) (3.67-4.84)
Insall-Burstein II
Microport[Fem] With 71 0.09 0.75 2.24 4.51 6.37 7.24
1,106 43
Insall-Burstein Patella (65 to 77) (0.01-0.65) (0.38-1.50) (1.49-3.35) (3.36-6.05) (4.89-8.27) (5.45-9.60)
(Microport)[Tib]
Without 71 0.66 2.94 3.76 5.91 7.90 8.36
914 48
Patella (65 to 77) (0.30-1.47) (2.01-4.29) (2.69-5.25) (4.49-7.75) (6.11-10.17) (6.42-10.85)
Journey II BCS
With 66 0.45 1.48 1.59
Patella (59 to 73) (0.26-0.75) (1.05-2.08) (1.12-2.24)
Journey[Tib]
1

(IQR)
Patella age at Male
Without 65 1.15 5.69 6.10
709 43
Patella (57 to 72) (0.58-2.29) (4.08-7.92) (4.36-8.51)
Kinemax With 71 0.25 1.24 1.76 3.68 5.56 5.98
4,381 37
[Fem:Tib] Patella (64 to 77) (0.14-0.46) (0.95-1.62) (1.41-2.21) (3.12-4.33) (4.80-6.44) (5.13-6.96)
Without 71 0.25 2.07 3.32 5.44 7.42 8.02
6,534 47
Patella (64 to 77) (0.15-0.40) (1.75-2.45) (2.90-3.80) (4.88-6.05) (6.71-8.20) (7.21-8.92)
LCS Complete[Fem] With 69 0.56 2.08 3.41 5.05 5.83
1,437 33
M.B.T.[Tib] Patella (62 to 76) (0.28-1.12) (1.43-3.02) (2.52-4.62) (3.83-6.64) (4.31-7.87)
Without 70 0.42 1.67 2.45 3.58 4.15
27,702 45
Patella (63 to 76) (0.35-0.50) (1.52-1.83) (2.27-2.65) (3.33-3.84) (3.82-4.50)
With 69.5 1.36 4.64 5.13 5.66 7.15 7.15
LCS[Fem:Tib] 220 37
Patella (63 to 76) (0.44-4.17) (2.52-8.45) (2.87-9.07) (3.25-9.77) (4.25-11.91) (4.25-11.91)
Without 70 0.57 1.43 1.97 2.67 3.38 3.70
1,781 42
Patella (63 to 76) (0.30-1.05) (0.97-2.11) (1.41-2.75) (1.99-3.57) (2.57-4.44) (2.81-4.87)
Legion CR COCR[Fem] With 69 1.18 2.41 3.08
170 34
Genesis II[Tib] Patella (62 to 76) (0.30-4.65) (0.91-6.30) (1.29-7.25)
Without 71 0.34 1.46 1.95
875 46
Patella (66 to 78) (0.11-1.06) (0.83-2.56) (1.17-3.23)
With 71 0.59 1.62 2.26 4.95 7.32
Maxim[Fem:Tib] 513 33
Patella (63 to 76) (0.19-1.83) (0.81-3.21) (1.26-4.04) (3.19-7.64) (4.85-10.96)
Without 70 0.33 1.84 2.98 5.70 9.62 12.46
1,231 47

Patella (63 to 77) (0.12-0.88) (1.21-2.78) (2.15-4.13) (4.46-7.28) (7.59-12.17) (8.68-17.71)
With 71 0.27 1.06 1.55 2.46 2.99 2.99
MRK[Fem:Tib] 5,325 38
Patella (64 to 77) (0.16-0.45) (0.80-1.40) (1.22-1.96) (1.96-3.08) (2.23-4.00) (2.23-4.00)
Without 70 0.34 1.23 1.66 2.84 3.29
9,793 48
Patella (64 to 76) (0.24-0.48) (1.02-1.49) (1.41-1.97) (2.41-3.36) (2.67-4.06)
Natural Knee II[Fem] With 70 0.46 1.66 2.65 4.34 7.83
1,531 41
NK2[Tib] Patella (64 to 76) (0.22-0.96) (1.13-2.45) (1.94-3.60) (3.37-5.59) (5.68-10.75)
Without 70 0.16 0.96 1.70 3.62 6.00 6.63
1,283 42
Patella (63 to 76) (0.04-0.63) (0.55-1.68) (1.11-2.60) (2.67-4.91) (4.46-8.05) (4.79-9.14)
Nexgen LCCK[Fem] With 71 0.40 1.75 1.75 4.89
515 37
Nexgen[Tib] Patella (63 to 78) (0.10-1.60) (0.83-3.67) (0.83-3.67) (2.16-10.86)
Without 72 1.78 3.44 4.65 5.11 11.24
576 36
Patella (64 to 79) (0.96-3.29) (2.18-5.42) (3.05-7.06) (3.36-7.75) (4.73-25.44)
Nexgen With 70 0.41 1.34 2.14 3.72 4.77 5.11
51,137 37
[Fem:Tib] Patella (63 to 76) (0.36-0.47) (1.24-1.45) (2.00-2.29) (3.49-3.96) (4.39-5.19) (4.51-5.79)
Without 70 0.36 1.27 1.99 3.34 4.47 4.87
122,912 44
Patella (64 to 76) (0.33-0.40) (1.21-1.34) (1.90-2.08) (3.20-3.48) (4.21-4.75) (4.50-5.27)
Nexgen[Fem]
LPS (Legacy With 67 0.47 2.38 3.31 6.28 8.65 8.65
1,077 37
Posterior Stabilised Patella (59 to 74) (0.20-1.13) (1.60-3.53) (2.35-4.65) (4.79-8.20) (6.40-11.64) (6.40-11.64)
ZimmerBiomet)[Tib]
Without 67 0.47 1.67 2.27 3.49 4.87 4.87
2,162 51
Patella (59 to 75) (0.25-0.86) (1.20-2.32) (1.71-3.03) (2.70-4.50) (3.64-6.52) (3.64-6.52)
Nexgen[Fem] With 62 0.73 2.51 3.30 5.55 7.07
415 55
TM Monoblock[Tib] Patella (56 to 69) (0.24-2.24) (1.36-4.61) (1.93-5.63) (3.60-8.52) (4.46-11.12)
Without 64 0.60 2.66 3.33 4.25 4.75 4.75
3,858 57
Patella (58 to 71) (0.40-0.90) (2.19-3.22) (2.80-3.97) (3.63-4.98) (4.02-5.61) (4.02-5.61)
Optetrak CR[Fem] With 70 0.94 2.40 3.76 7.00 7.41
646 43
Optetrak[Tib] Patella (64 to 76) (0.42-2.08) (1.45-3.94) (2.51-5.61) (5.06-9.64) (5.36-10.20)
1

(IQR)
Patella age at Male
Without 69 0.81 4.12 5.64 8.74 10.26
993 43
Patella (63 to 76) (0.41-1.62) (3.04-5.58) (4.35-7.30) (6.99-10.90) (7.55-13.85)
Persona CR[Fem] With 69 0.32 0.65 0.93
1,865 41
Persona[Tib] Patella (62 to 75) (0.13-0.77) (0.27-1.55) (0.40-2.18)
Without 70 0.20 0.79 2.00
2,857 49
Patella (63 to 76) (0.08-0.49) (0.40-1.53) (1.04-3.85)
Persona PS[Fem] With 69 0.41 1.08 2.48
552 36
Persona[Tib] Patella (62 to 75) (0.10-1.64) (0.39-2.95) (1.14-5.33)
Without 70 0.46 1.93 3.36
894 46
Patella (64 to 76) (0.17-1.22) (1.14-3.26) (2.07-5.42)
With 65 0.45 1.70 2.39 3.50 4.66 5.09
Knee[Fem] 8,720 43
Patella (58 to 72) (0.33-0.62) (1.44-2.00) (2.08-2.74) (3.11-3.94) (3.98-5.46) (4.08-6.35)
M.B.T.[Tib]
Without 65 0.81 2.32 3.17 4.43 5.31 5.31
8,539 50
Patella (58 to 73) (0.64-1.03) (2.01-2.66) (2.81-3.58) (3.98-4.94) (4.68-6.03) (4.68-6.03)
With 71 0.36 1.09 1.55 2.17 2.85 3.14
Knee[Fem] 66,318 38
Patella (64 to 77) (0.32-0.41) (1.01-1.18) (1.45-1.65) (2.04-2.30) (2.67-3.04) (2.83-3.49)
PFC Bicondylar[Tib]
Without 70 0.41 1.41 1.93 2.73 3.57 3.82
104,082 46
Patella (64 to 76) (0.37-0.45) (1.34-1.49) (1.84-2.02) (2.62-2.85) (3.40-3.76) (3.59-4.08)
Knee[Fem] With 71 0.36 1.17 1.67 2.33 2.56

82,688 38
PFC Sigma Patella (64 to 77) (0.33-0.41) (1.10-1.26) (1.57-1.77) (2.19-2.47) (2.26-2.90)
Bicondylar[Tib]
Without 70 0.38 1.59 2.18 2.89
109,501 45
Patella (64 to 77) (0.34-0.41) (1.51-1.67) (2.09-2.28) (2.77-3.03)
With 73 1.37 4.13
Profix[Fem:Tib] 82 30 0 0
Patella (65 to 78) (0.19-9.33) (1.35-12.26)
Without 73 0.42 1.40 1.88 2.69 3.64 3.99
3,874 44
Patella (67 to 78) (0.26-0.68) (1.07-1.83) (1.49-2.37) (2.20-3.28) (2.84-4.64) (3.01-5.29)
With 69 0.86 2.70 3.52 6.12 8.11 8.11
Rotaglide +[Fem:Tib] 1,177 42
Patella (63 to 76) (0.46-1.59) (1.91-3.82) (2.59-4.77) (4.81-7.77) (6.47-10.16) (6.47-10.16)
Without 71 0.37 3.51 4.44 7.01 9.09 9.09
822 45
Patella (64 to 77) (0.12-1.13) (2.44-5.04) (3.21-6.13) (5.37-9.11) (7.07-11.67) (7.07-11.67)
Rotaglide With 71 0.49 2.37 3.77 4.33 6.26
1,430 39
[Fem:Tib] Patella (63 to 77) (0.24-1.03) (1.69-3.32) (2.87-4.95) (3.32-5.62) (4.60-8.50)
Without 67 5.26 5.26 5.26
19 37
Patella (60 to 75) (0.76-31.88) (0.76-31.88) (0.76-31.88)
With 69 0.54 0.72 0.95
Saiph[Fem:Tib] 1,025 33
Patella (62 to 75) (0.22-1.29) (0.32-1.63) (0.43-2.06)
Without 70 0.56 2.03 2.03
830 47
Patella (63 to 76) (0.21-1.49) (1.14-3.59) (1.14-3.59)
With 71 0.45 1.30 1.99 3.13
Scorpio NRG[Fem:Tib] 7,127 39
Patella (64 to 77) (0.32-0.64) (1.06-1.59) (1.68-2.35) (2.69-3.64)
Without 70 0.37 1.89 2.88 4.25
6,974 46
Patella (64 to 76) (0.25-0.55) (1.60-2.25) (2.50-3.31) (3.75-4.82)
With 68 0.21 1.71 2.37 3.85 5.16 5.16
Scorpio[Fem:Tib] 959 40
Patella (60 to 75) (0.05-0.84) (1.05-2.77) (1.57-3.57) (2.76-5.36) (3.50-7.59) (3.50-7.59)
1

(IQR)
Patella age at Male
Without 68 0.44 2.37 3.44 5.04 6.36 6.36
2,296 47
Patella (62 to 75) (0.24-0.81) (1.81-3.09) (2.76-4.29) (4.18-6.07) (5.19-7.77) (5.19-7.77)
Scorpio[Fem] With 71 0.32 1.35 2.05 3.27 4.19 4.30
8,115 38
Scorpio NRG[Tib] Patella (65 to 77) (0.22-0.47) (1.12-1.63) (1.76-2.39) (2.88-3.70) (3.72-4.73) (3.79-4.88)
Without 71 0.51 2.12 2.98 4.47 5.73 5.98
13,574 44
Patella (64 to 77) (0.41-0.65) (1.89-2.38) (2.70-3.28) (4.12-4.85) (5.27-6.24) (5.42-6.60)
Sphere[Fem] With 69 0.59 1.32 2.18
387 36
GMK[Tib] Patella (61 to 75) (0.15-2.36) (0.49-3.49) (0.82-5.69)
Without 69 0.87 2.29 3.18
1,311 47
Patella (62 to 75) (0.48-1.56) (1.54-3.39) (2.18-4.61)

With 71 0.34 1.38 2.36 3.79 5.17 6.34
TC Plus[Fem:Tib] 890 37
Patella (64 to 76) (0.11-1.05) (0.79-2.42) (1.53-3.64) (2.63-5.43) (3.61-7.38) (3.98-10.01)
Without 70 0.70 1.82 2.38 3.51 4.79 5.11
15,140 45
Patella (64 to 76) (0.58-0.85) (1.61-2.05) (2.14-2.64) (3.21-3.83) (4.32-5.32) (4.47-5.84)
Triathlon With 70 0.49 1.27 1.78 2.67
64,214 39
[Fem:Tib] Patella (63 to 76) (0.44-0.54) (1.18-1.37) (1.67-1.91) (2.48-2.89)
Without 70 0.50 1.62 2.27 3.33 4.61
80,842 46
Patella (63 to 76) (0.45-0.55) (1.53-1.72) (2.15-2.40) (3.13-3.55) (3.75-5.67)
Unity Knee[Fem] With 70 0.28 0.93 1.22
1,123 43
Unity[Tib] Patella (63 to 76) (0.09-0.85) (0.48-1.79) (0.67-2.22)
Without 69 0.31 0.31 0.31
335 52
Patella (62 to 75) (0.04-2.19) (0.04-2.19) (0.04-2.19)
Vanguard With 70 0.37 1.09 1.63 2.63
34,878 37
[Fem:Tib] Patella (63 to 76) (0.31-0.45) (0.98-1.22) (1.48-1.79) (2.23-3.10)
Without 70 0.39 1.68 2.33 3.26
47,624 45
Patella (63 to 76) (0.34-0.46) (1.56-1.81) (2.18-2.48) (3.01-3.52)
Vanguard[Fem] With 68 0.30 0.54 1.04 2.63
688 34
Maxim[Tib] Patella (61 to 75) (0.07-1.18) (0.17-1.72) (0.42-2.54) (1.39-4.94)
1
Tables 3.K7 (a) and (b) and Table 3.K8 show the
Kaplan-Meier estimates of the cumulative percentage
probability of first revision, for any indication, of a
primary TKR (Tables 3.K7 (a) and (b)) and primary UKR
(Table 3.K8) by implant brand.
Table 3.K8 KM estimates of cumulative revision (95% CI) by unicompartmental knee replacement brands.
(IQR)
age at Male
Brand1 N primary (%) 1 year 3 years 5 years 10 years 15 years 17 years
All
63 1.03 3.77 5.80 11.04 16.59 18.42
unicompartmental 141,094 50
(56 to 71) (0.98-1.09) (3.67-3.88) (5.67-5.94) (10.81-11.27) (16.16-17.04) (17.74-19.13)
knee replacements
Unicondylar
AMC/Uniglide 64 2.36 6.06 7.69 12.60 17.04
3,011 51
[Fem:Tib] (57 to 72) (1.88-2.97) (5.25-6.98) (6.78-8.71) (11.35-13.97) (15.17-19.11)
Journey Uni
62 1.31 3.53 5.46
(56 to 69) (0.83-2.08) (2.58-4.82) (4.08-7.29)
Journey Uni[Tib]
63 0.84 4.01 6.07 10.29 13.36 13.69
MG Uni[Fem:Tib] 2,262 55
(57 to 70) (0.54-1.32) (3.28-4.91) (5.16-7.15) (9.08-11.66) (11.85-15.06) (12.07-15.52)
Oxford Cementless 65 1.18 2.38 3.37 5.85
24,975 56
Partial Knee[Fem:Tib] (58 to 71) (1.05-1.33) (2.18-2.59) (3.10-3.67) (5.10-6.72)
Oxford Cementless
Partial Knee[Fem] 66 1.16 3.73 5.32 9.22
1,895 46
Oxford Partial (57 to 73) (0.76-1.78) (2.90-4.79) (4.26-6.63) (7.48-11.34)
Knee[Tib]
Oxford Single Peg

Cemented Partial
64 1.23 4.38 6.52 11.72 17.28 19.34
Knee[Fem] 43,046 52
(58 to 71) (1.13-1.34) (4.18-4.58) (6.28-6.76) (11.39-12.07) (16.71-17.87) (18.42-20.29)
Oxford Partial
Knee[Tib]
Oxford Twin Peg
Cemented Partial
65 0.82 2.47 3.71 7.01
Knee[Fem] 5,401 48
(57 to 72) (0.61-1.11) (2.06-2.96) (3.17-4.35) (5.95-8.26)
Oxford Partial
Knee[Tib]
Persona Partial 65 0.21 0.72
2,631 58
Knee[Fem:Tib] (58 to 72) (0.09-0.50) (0.36-1.42)
*Physica ZUK 63 0.33 1.86 2.91 5.87 8.68
19,083 55
[Fem:Tib] (56 to 70) (0.26-0.43) (1.66-2.08) (2.63-3.22) (5.25-6.57) (6.02-12.41)
Preservation 63 2.56 8.09 11.63 17.79 23.54 24.65
1,487 55
[Fem:Tib] (56 to 69) (1.87-3.51) (6.80-9.60) (10.09-13.39) (15.90-19.88) (21.26-26.02) (22.10-27.43)
Sigma HP (Uni)[Fem] 63 0.74 2.89 4.07 6.43
12,787 58
Sigma HP[Tib] (56 to 70) (0.60-0.90) (2.59-3.23) (3.69-4.48) (5.76-7.17)
Triathlon Uni[Fem] 62 1.21 4.26 6.86 8.76
1,518 56
Triathlon[Tib] (55 to 69) (0.77-1.92) (3.28-5.51) (5.48-8.58) (6.98-10.95)
Patellofemoral
58 0.69 4.18 7.31 14.74 21.92 23.11
Avon[Fem] 6,378 22
(50 to 67) (0.51-0.92) (3.70-4.73) (6.65-8.04) (13.68-15.88) (20.13-23.83) (20.89-25.52)
59 0.91 7.04 10.31 19.31
FPV[Fem] 1,649 23
(52 to 68) (0.55-1.51) (5.89-8.40) (8.91-11.90) (17.24-21.59)
Journey PFJ 58 1.79 7.34 12.59 21.97
2,187 23
Oxinium[Fem] (50 to 67) (1.30-2.45) (6.26-8.59) (11.11-14.24) (19.80-24.33)
58 2.70 9.36 13.71 25.24
Sigma HP (PF)[Fem] 1,302 23
(50 to 66) (1.94-3.74) (7.89-11.08) (11.91-15.75) (22.14-28.70)
56 0.61 4.48 7.13 14.26
Zimmer PFJ[Fem] 3,224 23
(49 to 65) (0.39-0.96) (3.75-5.35) (6.12-8.30) (12.03-16.88)
*Denotes that this brand is now marketed by Lima.

1
Table 3.K9 (a) shows Kaplan-Meier estimates of the bearing types which were implanted on at least 1,000
cumulative percentage probability of first revision occasions for UKR and 2,500 occasions for TKR.
of a primary TKR or primary UKR by implant brand Patient summaries of age and gender by brand are
and bearing / constraint type for those brands / also given.
Table 3.K9 (a) KM estimates of cumulative revision (95% CI) by fixation, constraint and brand.
Median
Time since primary
(IQR)
age at Male
Total knee replacements
AGC V2[Fem:Tib]
Cemented, 71 0.26 1.42 2.08 3.33 5.42 6.27
37,070 43
unconstrained, fixed (65 to 77) (0.22-0.32) (1.30-1.55) (1.94-2.24) (3.13-3.53) (5.05-5.82) (5.67-6.92)
AGC[Fem]AGC V2[Tib]
Cemented, 71 0.31 1.58 2.22 3.51 5.80 6.32
28,092 42
Advance MP[Fem]Advance[Tib]
Cemented, 70 0.56 2.03 2.83 4.14 4.90 6.38
8,748 48
Attune CR[Fem]Attune FB[Tib]
Cemented, 69 0.37 1.43 1.90
18,550 44
unconstrained, fixed (62 to 75) (0.29-0.47) (1.25-1.64) (1.65-2.18)

Attune CR[Fem]Attune RP[Tib]
Cemented, 70 0.15 0.88 1.47
3,493 42
unconstrained, mobile (63 to 77) (0.06-0.36) (0.57-1.36) (0.93-2.33)
Attune PS[Fem]Attune FB[Tib]
Cemented, posterior- 70 0.45 1.62 2.48
10,159 42
stabilised, fixed (63 to 76) (0.34-0.61) (1.36-1.93) (2.07-2.99)
Columbus Cemented[Fem]Columbus CR/PS[Tib]
Cemented, 70 0.43 1.49 2.06 2.97 3.72
13,101 44
unconstrained, fixed (64 to 76) (0.33-0.56) (1.29-1.72) (1.81-2.35) (2.58-3.41) (3.04-4.56)
Genesis II Oxinium[Fem]Genesis II[Tib]
Cemented, 59 0.56 2.11 3.07 5.03 6.67
7,654 40
Cemented, posterior- 58 0.65 3.19 4.78 8.37 9.87
3,454 40
stabilised, fixed (53 to 64) (0.43-0.98) (2.63-3.87) (4.06-5.62) (7.22-9.69) (8.20-11.85)
Genesis II[Fem:Tib]
Cemented, 71 0.40 1.35 1.84 2.71 3.05 3.49
62,064 43
Cemented, posterior- 71 0.64 1.82 2.51 3.67 5.06
21,688 39
stabilised, fixed (65 to 77) (0.54-0.75) (1.64-2.02) (2.28-2.75) (3.32-4.04) (3.54-7.20)
Journey II BCS Oxinium[Fem]Journey[Tib]
Cemented, posterior- 66 0.58 2.40 2.57
4,049 41
stabilised, fixed (59 to 73) (0.38-0.88) (1.88-3.05) (2.01-3.28)

1
Brands shown have been used in at least 2,500 total primary knee replacement operations for that type of fixation and bearing type and at least 1,000 for
unicondylar and patellofemoral knee replacement operations.
Median
Time since primary
(IQR)
age at Male
Kinemax[Fem:Tib]
Cemented, 71 0.24 1.74 2.68 4.71 6.59 7.12
10,766 43
LCS Complete[Fem]M.B.T.[Tib]
Cemented, 70 0.41 1.51 2.48 3.94 4.41
12,390 41
unconstrained, mobile (64 to 76) (0.31-0.54) (1.31-1.75) (2.21-2.79) (3.56-4.36) (3.96-4.90)
Uncemented, 69 0.42 1.83 2.53 3.43 4.11
15,831 47
MRK[Fem:Tib]
Cemented, 70 0.31 1.16 1.61 2.69 3.22 3.22
14,877 44
Natural Knee II[Fem]NK2[Tib]
Cemented, 70 0.34 1.41 2.20 3.88 6.65 7.12
2,684 41
Nexgen[Fem:Tib]
Cemented, 70 0.31 1.03 1.52 2.41 3.19 3.34
89,051 43
Cemented, posterior- 70 0.45 1.57 2.57 4.42 5.76 6.23

82,171 41
stabilised, fixed (64 to 77) (0.41-0.50) (1.49-1.66) (2.45-2.69) (4.24-4.61) (5.45-6.08) (5.79-6.70)
Nexgen[Fem]TM Monoblock[Tib]
Uncemented, 64 0.61 2.62 3.33 4.39 4.98 4.98
4,002 58
PFC Sigma Bicondylar Knee[Fem]M.B.T.[Tib]
Cemented, 64 0.59 1.91 2.66 3.84 5.25 5.66
8,377 47
unconstrained, mobile (58 to 72) (0.45-0.78) (1.64-2.24) (2.33-3.04) (3.42-4.31) (4.49-6.14) (4.62-6.91)
7,135 46
stabilised, mobile (59 to 72) (0.50-0.88) (1.86-2.55) (2.64-3.45) (3.75-4.76) (4.21-5.45) (4.21-5.45)
PFC Sigma Bicondylar Knee[Fem]PFC Bicondylar[Tib]
Cemented, 70 0.39 1.23 1.70 2.36 3.02 3.19
132,331 43
36,344 41
PFC Sigma Bicondylar Knee[Fem]PFC Sigma Bicondylar[Tib]
Cemented, 70 0.35 1.34 1.87 2.50
122,269 42
unconstrained, fixed (63 to 76) (0.32-0.38) (1.27-1.41) (1.79-1.95) (2.39-2.62)
Cemented, posterior- 71 0.42 1.61 2.22 3.08
55,012 41
stabilised, fixed (64 to 77) (0.37-0.48) (1.50-1.72) (2.09-2.36) (2.90-3.27)
Cemented, monobloc 74 0.35 1.30 1.72 2.10 2.21
14,312 42
polyethylene tibia (69 to 79) (0.27-0.46) (1.11-1.51) (1.50-1.97) (1.83-2.42) (1.88-2.60)
Persona CR[Fem]Persona[Tib]
Cemented, 70 0.26 0.77 1.59
4,404 46
unconstrained, fixed (63 to 76) (0.14-0.49) (0.45-1.30) (0.89-2.85)

1
Median
Time since primary
(IQR)
age at Male
Scorpio NRG[Fem:Tib]
Cemented, 70 0.36 1.45 2.36 3.55
8,584 42
4,734 43
Scorpio[Fem]Scorpio NRG[Tib]
Cemented, 71 0.44 1.85 2.58 3.89 5.13 5.21
10,455 42
Cemented, posterior- 71.5 0.22 1.67 2.58 4.17 5.35 5.77
6,058 40
Uncemented, 70 0.62 1.93 2.61 3.93 4.74 4.74
3,733 47
TC Plus[Fem:Tib]
Cemented, 70 0.81 2.01 2.63 3.75 4.89 5.51
7,929 46

Cemented, 70 0.53 1.55 2.09 3.25 4.29
5,266 44
Triathlon[Fem:Tib]
Cemented, 70 0.46 1.39 1.93 2.89 4.24
113,969 43
24,445 41
Uncemented, 68 0.61 1.70 2.15 3.07
4,190 51
Vanguard[Fem:Tib]
Cemented, 70 0.35 1.36 1.95 2.86
67,485 42
10,208 40
Cemented, 70 0.47 1.28 1.56
3,380 36
constrained condylar (63 to 76) (0.28-0.77) (0.92-1.78) (1.14-2.13)
Unicondylar knee replacements
AMC/Uniglide[Fem:Tib]
1,087 50
Journey Uni Oxinium[Fem]Journey Uni[Tib]
62 1.47 3.29 4.75
Cemented, fixed 1,314 54
(56 to 69) (0.93-2.32) (2.34-4.62) (3.41-6.58)
MG Uni[Fem:Tib]
62 0.95 4.36 6.59 11.45 14.36 14.36
(56 to 69) (0.56-1.59) (3.43-5.54) (5.43-7.99) (9.89-13.24) (12.45-16.54) (12.45-16.54)
Oxford Cementless Partial Knee[Fem:Tib]
Uncemented/Hybrid, 65 1.18 2.38 3.37 5.85
24,975 56
mobile (58 to 71) (1.05-1.33) (2.18-2.59) (3.10-3.67) (5.10-6.72)

1
Median
Time since primary
(IQR)
age at Male
Oxford Cementless Partial Knee[Fem]Oxford Partial Knee[Tib]
1,496 50
mobile (58 to 73) (0.94-2.19) (3.27-5.42) (4.63-7.22) (7.88-11.86)
Oxford Single Peg Cemented Partial Knee[Fem]Oxford Partial Knee[Tib]
64 1.23 4.38 6.52 11.72 17.28 19.34
Cemented, mobile 43,021 52
(58 to 71) (1.13-1.33) (4.18-4.58) (6.28-6.76) (11.39-12.07) (16.71-17.88) (18.42-20.29)
Oxford Twin Peg Cemented Partial Knee[Fem]Oxford Partial Knee[Tib]
65 0.80 2.45 3.72 7.02
(57 to 72) (0.59-1.09) (2.04-2.95) (3.17-4.37) (5.95-8.28)
Persona Partial Knee[Fem:Tib]
65 0.21 0.72
(58 to 72) (0.09-0.50) (0.36-1.42)
*Physica ZUK[Fem:Tib]
63 0.35 1.71 2.74 5.61 8.48
(56 to 70) (0.27-0.45) (1.51-1.95) (2.45-3.06) (4.96-6.36) (5.82-12.28)
Cemented, monobloc 64 0.21 2.93 4.14 7.61

2,005 55
polyethylene tibia (56 to 71) (0.08-0.55) (2.24-3.83) (3.27-5.24) (5.94-9.73)
Sigma HP (Uni)[Fem]Sigma HP[Tib]
63 0.75 2.82 3.94 6.30
(56 to 70) (0.61-0.92) (2.52-3.15) (3.56-4.35) (5.61-7.07)
Triathlon Uni[Fem]Triathlon[Tib]
62 1.21 4.26 6.86 8.76
(55 to 69) (0.77-1.92) (3.28-5.51) (5.48-8.58) (6.98-10.95)
Patellofemoral knee replacements
Avon[Fem]
58 0.69 4.18 7.31 14.74 21.92 23.11
Patellofemoral 6,378 22
(50 to 67) (0.51-0.92) (3.70-4.73) (6.65-8.04) (13.68-15.88) (20.13-23.83) (20.89-25.52)
FPV[Fem]
59 0.91 7.04 10.31 19.31
(52 to 68) (0.55-1.51) (5.89-8.40) (8.91-11.90) (17.24-21.59)
Journey PFJ Oxinium[Fem]
58 1.79 7.34 12.59 21.97
(50 to 67) (1.30-2.45) (6.26-8.59) (11.11-14.24) (19.80-24.33)
Sigma HP (PF)[Fem]
58 2.70 9.36 13.71 25.24
(50 to 66) (1.94-3.74) (7.89-11.08) (11.91-15.75) (22.14-28.70)
Zimmer PFJ[Fem]
56 0.61 4.48 7.13 14.26
(49 to 65) (0.39-0.96) (3.75-5.35) (6.12-8.30) (12.03-16.88)

1
Table 3.K9 (b) KM estimates of cumulative revision (95% CI) by fixation, constraint, brand and whether a
patella component was recorded. Blue italics signify that fewer than 250 cases remained at risk at these time points.
Median
Time since primary
(IQR)
age at Male
Total knee replacements
AGC V2[Fem:Tib]
Cemented,
71 0.23 1.21 1.81 2.96 4.63 4.80
unconstrained, fixed, 11,769 35
(65 to 77) (0.16-0.34) (1.03-1.43) (1.58-2.08) (2.64-3.32) (4.03-5.32) (4.13-5.59)
with patella
Cemented,
71 0.28 1.52 2.21 3.49 5.74 6.82
(65 to 77) (0.22-0.35) (1.37-1.68) (2.03-2.40) (3.25-3.75) (5.29-6.23) (6.06-7.67)
without patella
AGC[Fem]AGC V2[Tib]
Cemented,
71 0.26 1.20 1.70 2.99 6.16 6.16
(64 to 77) (0.17-0.38) (1.00-1.45) (1.45-1.99) (2.61-3.42) (5.16-7.35) (5.16-7.35)
with patella
Cemented,
71 0.34 1.77 2.48 3.77 5.37 6.32
(64 to 77) (0.26-0.43) (1.59-1.97) (2.26-2.72) (3.47-4.10) (4.78-6.03) (5.21-7.65)
without patella
Advance MP[Fem]Advance[Tib]
Cemented,
70 0.50 1.48 2.03 3.28 3.90
(63 to 76) (0.30-0.83) (1.10-1.99) (1.56-2.63) (2.61-4.13) (2.97-5.13)
with patella

Cemented,
70 0.60 2.32 3.25 4.56 5.42 8.57
(64 to 76) (0.43-0.83) (1.95-2.75) (2.81-3.77) (3.97-5.24) (4.60-6.39) (4.15-17.26)
without patella
Attune CR[Fem]Attune FB[Tib]
Cemented,
70 0.24 1.13 1.54
(62 to 76) (0.15-0.39) (0.88-1.46) (1.20-1.97)
with patella
Cemented,
69 0.46 1.62 2.12
(62 to 75) (0.35-0.60) (1.38-1.91) (1.79-2.51)
without patella
Attune CR[Fem]Attune RP[Tib]
Cemented,
70 0.19 0.86 1.32
unconstrained, 2,140 37
(63 to 77) (0.07-0.52) (0.49-1.50) (0.70-2.49)
mobile, with patella
Cemented,
70 0.08 0.89 1.64
(63 to 77) (0.01-0.59) (0.44-1.80) (0.85-3.18)
mobile, without patella
Attune PS[Fem]Attune FB[Tib]
Cemented, posterior-
70 0.47 1.36 2.22
stabilised, fixed, with 6,183 41
(63 to 76) (0.33-0.69) (1.07-1.74) (1.69-2.90)
patella
70 0.42 2.03 2.92
stabilised, fixed, 3,976 44
(62 to 76) (0.26-0.68) (1.57-2.61) (2.27-3.76)
without patella
Columbus Cemented[Fem]Columbus CR/PS[Tib]
Cemented,
70 0.58 1.30 1.57 3.02 5.76
(63 to 76) (0.38-0.88) (0.98-1.73) (1.20-2.05) (2.08-4.37) (3.34-9.84)
with patella
Cemented,
71 0.37 1.56 2.25 3.05 3.35
(65 to 76) (0.27-0.52) (1.32-1.85) (1.94-2.61) (2.62-3.55) (2.82-3.99)
without patella

1
Median
Time since primary
(IQR)
age at Male
Genesis II Oxinium[Fem]Genesis II[Tib]
Cemented,
59 0.50 1.55 2.06 3.61 5.02
(54 to 64) (0.32-0.76) (1.20-1.99) (1.64-2.59) (2.92-4.44) (3.72-6.75)
with patella
Cemented,
59 0.64 2.81 4.31 6.74 8.59
(54 to 65) (0.42-0.98) (2.27-3.46) (3.61-5.13) (5.73-7.92) (7.15-10.31)
without patella
59 0.54 2.36 3.21 5.94
(54 to 65) (0.28-1.04) (1.71-3.27) (2.41-4.27) (4.52-7.78)
patella
57 0.75 3.95 6.20 10.46
(52 to 63) (0.43-1.28) (3.10-5.02) (5.09-7.55) (8.79-12.42)
without patella
Genesis II[Fem:Tib]
Cemented,
71 0.39 1.04 1.37 2.03 2.36 2.90
(66 to 77) (0.32-0.47) (0.92-1.18) (1.23-1.54) (1.82-2.28) (2.02-2.76) (1.97-4.27)
with patella
Cemented,
71 0.41 1.59 2.20 3.22 3.56 3.84
(65 to 77) (0.35-0.49) (1.45-1.73) (2.03-2.38) (2.97-3.48) (3.23-3.92) (3.24-4.55)
without patella
71 0.63 1.73 2.24 3.27 4.07

(65 to 77) (0.50-0.80) (1.49-2.01) (1.95-2.57) (2.83-3.79) (3.10-5.34)
patella
71 0.64 1.92 2.77 4.02 5.66
(65 to 77) (0.50-0.82) (1.66-2.22) (2.44-3.15) (3.53-4.58) (3.57-8.92)
without patella
Journey II BCS Oxinium[Fem]Journey[Tib]
66 0.45 1.48 1.59
(59 to 73) (0.26-0.75) (1.05-2.08) (1.13-2.24)
patella
65 1.15 5.55 5.96
stabilised, fixed, 707 43
(57 to 72) (0.58-2.29) (3.96-7.77) (4.24-8.37)
without patella
Kinemax[Fem:Tib]
Cemented,
71 0.26 1.24 1.75 3.64 5.47 5.89
(64 to 77) (0.14-0.47) (0.95-1.63) (1.39-2.20) (3.08-4.29) (4.72-6.34) (5.05-6.87)
with patella
Cemented,
71 0.23 2.07 3.30 5.41 7.33 7.94
(64 to 77) (0.14-0.39) (1.75-2.46) (2.88-3.78) (4.86-6.03) (6.62-8.11) (7.13-8.83)
without patella
LCS Complete[Fem]M.B.T.[Tib]
Cemented,
70 0.64 2.19 3.80 6.37 6.94
unconstrained, 781 31
(63 to 77) (0.27-1.54) (1.34-3.55) (2.60-5.54) (4.59-8.81) (4.94-9.71)
Cemented,
71 0.39 1.47 2.40 3.79 4.25
(64 to 76) (0.29-0.52) (1.26-1.71) (2.12-2.71) (3.40-4.21) (3.80-4.75)
Uncemented,
68 0.54 1.87 2.69 3.01 4.19
unconstrained, 567 34
(61 to 74) (0.17-1.66) (0.97-3.57) (1.52-4.72) (1.74-5.19) (2.13-8.16)

1

Median
Time since primary
(IQR)
age at Male
Uncemented,
69 0.42 1.83 2.52 3.44 4.10
(62 to 75) (0.33-0.53) (1.62-2.06) (2.27-2.80) (3.13-3.79) (3.65-4.60)
MRK[Fem:Tib]
Cemented,
71 0.25 1.01 1.50 2.42 2.95 2.95
(64 to 77) (0.15-0.43) (0.76-1.34) (1.18-1.92) (1.92-3.05) (2.20-3.96) (2.20-3.96)
with patella
Cemented,
70 0.35 1.24 1.66 2.85 3.30
(63 to 76) (0.25-0.49) (1.03-1.50) (1.40-1.97) (2.41-3.37) (2.67-4.07)
without patella
Natural Knee II[Fem]NK2[Tib]
Cemented,
70 0.46 1.68 2.67 4.39 7.84
(64 to 76) (0.22-0.97) (1.14-2.47) (1.96-3.64) (3.40-5.64) (5.63-10.87)
with patella
Cemented,
70 0.17 1.05 1.59 3.21 5.45 6.21
(64 to 76) (0.04-0.69) (0.60-1.85) (1.01-2.52) (2.28-4.51) (3.90-7.58) (4.25-9.02)
without patella
Nexgen[Fem:Tib]
Cemented,
70 0.30 0.99 1.45 2.36 3.13 3.13
(63 to 76) (0.23-0.37) (0.86-1.13) (1.29-1.63) (2.09-2.66) (2.59-3.79) (2.59-3.79)

with patella
Cemented,
70 0.31 1.04 1.54 2.43 3.20 3.41
(63 to 76) (0.27-0.36) (0.96-1.13) (1.44-1.65) (2.27-2.60) (2.83-3.61) (2.89-4.03)
without patella
70 0.52 1.67 2.77 4.83 6.07 6.50
(63 to 76) (0.44-0.62) (1.52-1.85) (2.55-3.00) (4.49-5.20) (5.54-6.64) (5.71-7.40)
patella
71 0.42 1.52 2.48 4.24 5.63 6.12
(64 to 77) (0.37-0.48) (1.42-1.63) (2.34-2.63) (4.03-4.46) (5.25-6.03) (5.61-6.67)
without patella
Nexgen[Fem]TM Monoblock[Tib]
Uncemented,
63 0.53 2.21 3.09 5.20 6.87
unconstrained, fixed, 379 58
(57 to 69) (0.13-2.11) (1.11-4.38) (1.72-5.52) (3.25-8.27) (4.16-11.25)
with patella
Uncemented,
65 0.61 2.66 3.36 4.31 4.82 4.82
(58 to 72) (0.40-0.93) (2.18-3.25) (2.80-4.02) (3.66-5.07) (4.06-5.71) (4.06-5.71)
without patella
PFC Sigma Bicondylar Knee[Fem]M.B.T.[Tib]
Cemented,
64 0.47 2.12 2.88 4.32 6.34
(58 to 72) (0.29-0.78) (1.67-2.70) (2.34-3.53) (3.63-5.14) (5.02-7.99)
Cemented,
64 0.66 1.79 2.52 3.51 4.51
(58 to 71) (0.47-0.92) (1.46-2.19) (2.12-3.00) (3.01-4.11) (3.66-5.57)
64 0.45 1.45 2.10 2.98 3.46
stabilised, mobile, 5,153 45
(59 to 72) (0.30-0.67) (1.15-1.82) (1.73-2.53) (2.52-3.52) (2.82-4.23)
with patella
66 1.22 4.11 5.45 7.47 8.23
stabilised, mobile, 1,982 49
(58 to 73) (0.82-1.82) (3.31-5.09) (4.51-6.57) (6.31-8.82) (6.95-9.73)
without patella

1
Median
Time since primary
(IQR)
age at Male
PFC Sigma Bicondylar Knee[Fem]PFC Bicondylar[Tib]
Cemented,
71 0.35 1.02 1.48 2.04 2.68 2.79
(64 to 77) (0.30-0.41) (0.92-1.12) (1.36-1.60) (1.89-2.20) (2.45-2.92) (2.52-3.09)
with patella
Cemented,
70 0.41 1.34 1.80 2.51 3.19 3.39
(64 to 76) (0.37-0.45) (1.26-1.42) (1.71-1.90) (2.39-2.64) (3.02-3.38) (3.17-3.63)
without patella
71 0.40 1.23 1.68 2.39 3.13 3.71
(64 to 77) (0.32-0.49) (1.09-1.39) (1.51-1.87) (2.17-2.63) (2.83-3.47) (3.05-4.50)
patella
71 0.40 1.86 2.59 3.84 5.32 5.74
(64 to 77) (0.31-0.52) (1.65-2.09) (2.34-2.87) (3.51-4.20) (4.80-5.88) (4.93-6.67)
without patella
PFC Sigma Bicondylar Knee[Fem]PFC Sigma Bicondylar[Tib]
Cemented,
70 0.35 1.16 1.64 2.24
(63 to 76) (0.30-0.41) (1.05-1.27) (1.51-1.78) (2.05-2.45)
with patella
Cemented,
70 0.35 1.43 1.99 2.63
(63 to 76) (0.31-0.39) (1.35-1.52) (1.88-2.10) (2.49-2.78)
without patella
71 0.38 1.20 1.69 2.43
(65 to 77) (0.32-0.45) (1.09-1.32) (1.55-1.84) (2.23-2.65)
patella
70 0.51 2.38 3.21 4.26
(63 to 77) (0.42-0.62) (2.16-2.62) (2.95-3.50) (3.93-4.63)
without patella
Cemented, monobloc
76 0.41 1.10 1.61 1.80 2.10
polyethylene tibia, 2,785 37
(71 to 81) (0.23-0.73) (0.75-1.60) (1.16-2.24) (1.30-2.51) (1.41-3.13)
with patella
Cemented, monobloc
74 0.34 1.34 1.74 2.18
polyethylene tibia, 11,527 43
(69 to 79) (0.25-0.46) (1.14-1.58) (1.49-2.03) (1.87-2.55)
without patella
Persona CR[Fem]Persona[Tib]
Cemented,
69 0.33 0.66 0.66
(62 to 75) (0.14-0.80) (0.28-1.57) (0.28-1.57)
with patella
Cemented,
70 0.22 0.83 2.04
(63 to 76) (0.09-0.52) (0.42-1.60) (1.07-3.90)
without patella
Scorpio NRG[Fem:Tib]
Cemented,
70 0.42 1.23 2.01 3.39
(64 to 76) (0.26-0.69) (0.92-1.64) (1.60-2.53) (2.75-4.18)
with patella
Cemented,
70 0.31 1.63 2.63 3.71
(64 to 76) (0.19-0.52) (1.30-2.03) (2.20-3.14) (3.16-4.35)
without patella
71 0.49 1.31 1.90 2.85
(64 to 77) (0.29-0.80) (0.96-1.78) (1.47-2.46) (2.27-3.57)
patella
69 0.37 2.45 3.43 5.62
(63 to 76) (0.17-0.82) (1.80-3.34) (2.64-4.46) (4.52-6.96)
without patella

1

Median
Time since primary
(IQR)
age at Male
Scorpio[Fem]Scorpio NRG[Tib]
Cemented,
72 0.36 1.23 1.89 3.35 4.24 4.51
(65 to 77) (0.20-0.65) (0.89-1.70) (1.46-2.46) (2.73-4.10) (3.46-5.18) (3.60-5.63)
with patella
Cemented,
70 0.48 2.10 2.86 4.12 5.49 5.49
(64 to 77) (0.34-0.66) (1.79-2.46) (2.50-3.28) (3.67-4.62) (4.87-6.20) (4.87-6.20)
without patella
71 0.15 1.16 1.81 3.05 4.13 4.13
(65 to 77) (0.06-0.35) (0.85-1.58) (1.41-2.32) (2.50-3.72) (3.44-4.96) (3.44-4.96)
patella
72 0.31 2.36 3.61 5.69 7.00 7.89
(65 to 77) (0.16-0.63) (1.83-3.03) (2.94-4.43) (4.82-6.71) (5.97-8.20) (6.40-9.72)
without patella
Uncemented,
71 0.37 1.75 2.53 3.26 4.02
(63 to 77) (0.12-1.15) (1.04-2.94) (1.64-3.89) (2.21-4.80) (2.74-5.88)
with patella
Uncemented,
70 0.69 1.98 2.63 4.11 4.94 4.94
(64 to 76) (0.44-1.07) (1.53-2.56) (2.10-3.29) (3.42-4.94) (4.12-5.91) (4.12-5.91)
without patella
TC Plus[Fem:Tib]
Cemented,

71 0.18 1.45 2.58 3.91 5.43 5.43
(64 to 76) (0.03-1.27) (0.73-2.89) (1.53-4.31) (2.53-6.01) (3.56-8.26) (3.56-8.26)
with patella
Cemented,
70 0.86 2.05 2.64 3.74 4.83 5.62
(64 to 76) (0.67-1.10) (1.75-2.40) (2.29-3.03) (3.31-4.21) (4.22-5.53) (4.45-7.09)
without patella
Cemented,
72 0.47 1.47 1.47
unconstrained, 237 35 0
(65 to 77) (0.07-3.29) (0.48-4.49) (0.48-4.49)
Cemented,
70 0.56 1.60 2.12 3.31 4.37
(64 to 76) (0.39-0.81) (1.28-1.99) (1.75-2.56) (2.83-3.88) (3.69-5.17)
Triathlon[Fem:Tib]
Cemented,
70 0.44 1.17 1.59 2.46
(63 to 76) (0.38-0.50) (1.07-1.28) (1.46-1.74) (2.23-2.71)
with patella
Cemented,
70 0.47 1.53 2.15 3.16 4.43
(63 to 76) (0.42-0.53) (1.43-1.63) (2.02-2.28) (2.95-3.40) (3.54-5.53)
without patella
70 0.57 1.50 2.27 3.20
(63 to 76) (0.46-0.70) (1.31-1.72) (2.01-2.55) (2.83-3.61)
patella
70 0.71 2.23 3.13 4.56
(63 to 77) (0.55-0.92) (1.91-2.60) (2.72-3.60) (3.90-5.32)
without patella
Uncemented,
68 0.83 1.51 1.51 1.51
(60 to 75) (0.43-1.60) (0.85-2.68) (0.85-2.68) (0.85-2.68)
with patella
Uncemented,
69 0.53 1.76 2.28 3.26
(61 to 75) (0.32-0.88) (1.30-2.39) (1.69-3.06) (2.17-4.90)
without patella

1
Median
Time since primary
(IQR)
age at Male
Vanguard[Fem:Tib]
Cemented,
70 0.33 0.96 1.44 2.52
(64 to 76) (0.26-0.40) (0.84-1.09) (1.28-1.62) (2.05-3.09)
with patella
Cemented,
70 0.37 1.61 2.27 3.14
(63 to 76) (0.32-0.44) (1.49-1.75) (2.11-2.44) (2.88-3.41)
without patella
70 0.56 1.68 2.48 3.24
(63 to 76) (0.39-0.79) (1.35-2.08) (2.05-2.99) (2.64-3.98)
patella
70 0.68 2.67 3.36 5.11
(63 to 77) (0.48-0.97) (2.21-3.22) (2.82-4.01) (4.01-6.50)
without patella
Cemented,
70 0.54 1.19 1.60
constrained condylar, 1,732 32
(63 to 76) (0.28-1.03) (0.75-1.88) (1.05-2.45)
with patella
Cemented,
70 0.39 1.38 1.52
constrained condylar, 1,648 40
(63 to 76) (0.17-0.86) (0.87-2.19) (0.96-2.40)
without patella
Unicondylar knee replacements

AMC/Uniglide[Fem:Tib]
1,087 50
Journey Uni Oxinium[Fem]Journey Uni[Tib]
62 1.47 3.29 4.75
(56 to 69) (0.93-2.32) (2.34-4.62) (3.41-6.58)
MG Uni[Fem:Tib]
62 0.95 4.36 6.59 11.45 14.36 14.36
(56 to 69) (0.56-1.59) (3.43-5.54) (5.43-7.99) (9.89-13.24) (12.45-16.54) (12.45-16.54)
Oxford Cementless Partial Knee[Fem:Tib]
24,975 56
mobile (58 to 71) (1.05-1.33) (2.18-2.59) (3.10-3.67) (5.10-6.72)
Oxford Cementless Partial Knee[Fem]Oxford Partial Knee[Tib]
1,496 50
mobile (58 to 73) (0.94-2.19) (3.27-5.42) (4.63-7.22) (7.88-11.86)
Oxford Single Peg Cemented Partial Knee[Fem]Oxford Partial Knee[Tib]
64 1.23 4.38 6.52 11.72 17.28 19.34
(58 to 71) (1.13-1.33) (4.18-4.58) (6.28-6.76) (11.39-12.07) (16.71-17.88) (18.42-20.29)
Oxford Twin Peg Cemented Partial Knee[Fem]Oxford Partial Knee[Tib]
65 0.80 2.45 3.72 7.02
(57 to 72) (0.59-1.09) (2.04-2.95) (3.17-4.37) (5.95-8.28)
Persona Partial Knee[Fem:Tib]
65 0.21 0.72
(58 to 72) (0.09-0.50) (0.36-1.42)

1

Median
Time since primary
(IQR)
age at Male
*Physica ZUK[Fem:Tib]
63 0.35 1.71 2.74 5.61 8.48
(56 to 70) (0.27-0.45) (1.51-1.95) (2.45-3.06) (4.96-6.36) (5.82-12.28)
Cemented, monobloc 64 0.21 2.93 4.14 7.61
2,005 55
polyethylene tibia (56 to 71) (0.08-0.55) (2.24-3.83) (3.27-5.24) (5.94-9.73)

Sigma HP (Uni)[Fem]Sigma HP[Tib]
63 0.75 2.82 3.94 6.30
(56 to 70) (0.61-0.92) (2.52-3.15) (3.56-4.35) (5.61-7.07)
Triathlon Uni[Fem]Triathlon[Tib]
62 1.21 4.26 6.86 8.76
(55 to 69) (0.77-1.92) (3.28-5.51) (5.48-8.58) (6.98-10.95)
Patellofemoral knee replacements
Avon[Fem]
58 0.69 4.18 7.31 14.74 21.92 23.11
(50 to 67) (0.51-0.92) (3.70-4.73) (6.65-8.04) (13.68-15.88) (20.13-23.83) (20.89-25.52)
FPV[Fem]
59 0.91 7.04 10.31 19.31
(52 to 68) (0.55-1.51) (5.89-8.40) (8.91-11.90) (17.24-21.59)
Journey PFJ Oxinium[Fem]
58 1.79 7.34 12.59 21.97
(50 to 67) (1.30-2.45) (6.26-8.59) (11.11-14.24) (19.80-24.33)
Sigma HP (PF)[Fem]
58 2.70 9.36 13.71 25.24
(50 to 66) (1.94-3.74) (7.89-11.08) (11.91-15.75) (22.14-28.70)
Zimmer PFJ[Fem]
56 0.61 4.48 7.13 14.26
(49 to 65) (0.39-0.96) (3.75-5.35) (6.12-8.30) (12.03-16.88)

1
3.3.4 Revisions for different implant wear and progressive arthritis were lower for
uncemented / hybrid fixation than for cemented but
indications after primary knee the incidence was higher for dislocation / subluxation
replacement and periprosthetic fracture. For patellofemoral
replacements, the top three indications for revision
Table 3.K10 shows the revision incidence rates for
were: progressive arthritis, pain and ‘other’ indication.
each indication recorded on data collection forms
Similarly, for multicompartmental knee replacements,
for knee revision surgery, for all cases and then sub-
the highest incidence for revision was for progressive
divided by fixation type and whether the primary
arthritis, pain and ‘other’ indication.
procedure was a TKR or a UKR.
In Table 3.K11 (page 190), the PTIRs for each
For all knee replacements, the highest PTIRs for
indication are shown separately for different time
the five most common indications for revision in
periods from the primary knee replacement, within the
descending order, were for: aseptic loosening / lysis,
first year from primary operation, and between 1 to
infection, progressive arthritis, pain and instability.
3, 3 to 5, 5 to 7, 7 to 10, 10 to 13, 13 to 15, 15 to 17
For cemented TKR, the highest PTIRs in descending
and ≥17 years after surgery (the maximum follow-up
order were aseptic loosening / lysis, infection,
for any implant is now 17.75 years). It is clear that
instability, pain and ‘other’ indication. Revision
most of the PTIRs for a particular indication do vary,
incidences for pain and aseptic loosening / lysis, wear
especially for infection, aseptic loosening / lysis, pain
and ‘other’ indications were slightly higher for TKRs
and progressive arthritis for different time intervals
which were uncemented, compared to prosthesis
after surgery. Infection is most likely to be the reason
implanted using a cemented fixation, but revision for
that a joint is revised in the first year but after seven
infection was lower for uncemented.
years or more, is comparatively less likely than some
For cemented unicondylar knee replacements (medial of the other reasons. Conversely, revision between
and lateral UKR), the highest three incidence rates one and three years after surgery is more likely for
for indications for revising the implant were for: aseptic loosening / lysis and pain, with incidence
progressive arthritis, aseptic loosening / lysis and pain, rates dropping off for pain later on but rising again
respectively. For uncemented / hybrid unicondylar for aseptic loosening / lysis. Aseptic loosening / lysis
knee replacements (medial and lateral UKR) the PTIRs continue to remain relatively higher than other
highest rates were for: progressive arthritis, aseptic indicated reasons for revision for implants surviving for
loosening / lysis and dislocation / subluxation. The longer periods after surgery.
incidence of revision for pain, aseptic loosening / lysis,
Table 3.K10 PTIR estimates of indications for revision (95% CI) by fixation, constraint, bearing type and whether a patella component was recorded.
Number of revisions per 1,000 prosthesis-years for: Stiffness3 Progressive arthritis4

Prosthe- Revisions
sis- Prosthe- per 1,000 Prosthe- Revisions
Fixation, constraint years at Aseptic Peri- sis-years prosthesis-years per 1,000
and bearing sub- risk Dislocation/ loosening prosthetic Implant Malalign- Other at risk sis- at risk prosthesis-
groups (x1,000) All causes Pain Subluxation Infection / Lysis fracture wear1 Instability ment indication2 (x1,000) years (x1,000) years
4.53 0.68 0.16 0.87 1.21 0.17 0.29 0.64 0.32 0.50 0.29 0.73
All cases 8,921.4 8,665.1 6,503.2
(4.49-4.58) (0.66-0.70) (0.16-0.17) (0.85-0.89) (1.18-1.23) (0.16-0.18) (0.28-0.30) (0.63-0.66) (0.31-0.34) (0.48-0.51) (0.27-0.30) (0.71-0.75)
3.62 0.48 0.10 0.94 0.98 0.16 0.18 0.60 0.29 0.35 0.29 0.34
All cemented 7,353.0 7,155.1 5,453.7
(3.58-3.66) (0.47-0.50) (0.09-0.11) (0.92-0.96) (0.96-1.01) (0.15-0.17) (0.17-0.19) (0.59-0.62) (0.27-0.30) (0.33-0.36) (0.28-0.31) (0.33-0.36)
unconstrained, fixed, 2.86 0.31 0.08 0.90 0.82 0.13 0.18 0.57 0.26 0.22 0.26 0.03
1,708.7 1,657.4 1,287.0
with patella (2.78-2.94) (0.28-0.33) (0.07-0.09) (0.85-0.94) (0.78-0.87) (0.11-0.14) (0.16-0.20) (0.54-0.61) (0.23-0.28) (0.20-0.25) (0.23-0.28) (0.02-0.04)
3,249.2 3,168.7 2,438.1
without patella (3.48-3.61) (0.53-0.58) (0.08-0.10) (0.81-0.87) (0.78-0.84) (0.12-0.14) (0.15-0.17) (0.54-0.59) (0.27-0.30) (0.37-0.41) (0.28-0.32) (0.51-0.56)
unconstrained, 5.03 0.52 0.33 1.27 1.67 0.15 0.44 1.12 0.40 0.29 0.58
82.1 78.7 49.2 0
mobile, with patella (4.57-5.54) (0.39-0.71) (0.23-0.48) (1.05-1.54) (1.41-1.97) (0.08-0.26) (0.32-0.61) (0.91-1.38) (0.29-0.57) (0.20-0.44) (0.44-0.78)
unconstrained,
3.92 0.71 0.15 0.77 1.28 0.16 0.28 0.70 0.37 0.35 0.36 0.32
mobile, without 254.3 248.0 150.4
(3.69-4.18) (0.61-0.82) (0.11-0.21) (0.67-0.89) (1.15-1.43) (0.12-0.21) (0.22-0.35) (0.60-0.81) (0.31-0.46) (0.28-0.43) (0.30-0.45) (0.24-0.42)
patella
900.3 874.9 672.4
fixed, with patella (3.50-3.75) (0.29-0.36) (0.07-0.11) (1.12-1.26) (1.13-1.27) (0.20-0.26) (0.15-0.20) (0.54-0.64) (0.25-0.31) (0.22-0.28) (0.23-0.30) (0.02-0.04)
887.6 860.2 645.4
fixed, without patella (4.79-5.08) (0.57-0.68) (0.08-0.12) (1.01-1.15) (1.51-1.67) (0.20-0.27) (0.18-0.24) (0.67-0.78) (0.29-0.37) (0.45-0.54) (0.28-0.35) (0.58-0.71)
71.1 69.9 49.4 0
posterior-stabilised,
6.19 1.07 0.24 0.75 1.31 0.32 0.43 1.01 0.19 1.23 0.60 1.16
mobile, without 37.5 36.4 20.7
(5.44-7.04) (0.78-1.46) (0.12-0.46) (0.52-1.08) (0.99-1.73) (0.18-0.56) (0.26-0.70) (0.74-1.39) (0.09-0.39) (0.92-1.64) (0.40-0.92) (0.78-1.73)
patella
constrained condylar, 4.39 0.19 0.34 2.22 0.68 0.29 0.05 0.63 0.05 0.43 0.24 0.11
20.7 20.5 18.2
with patella (3.57-5.39) (0.07-0.51) (0.16-0.71) (1.66-2.96) (0.40-1.14) (0.13-0.64) (0.01-0.34) (0.36-1.08) (0.01-0.34) (0.23-0.83) (0.10-0.59) (0.03-0.44)
constrained condylar, 5.94 0.38 0.30 2.61 0.91 0.49 0.34 0.76 0.38 0.45 0.35 0.67
26.4 26.1 22.4
without patella (5.08-6.94) (0.20-0.70) (0.15-0.60) (2.06-3.30) (0.61-1.35) (0.29-0.85) (0.18-0.65) (0.49-1.17) (0.20-0.70) (0.26-0.80) (0.18-0.66) (0.40-1.11)
monobloc
2.37 0.26 0.04 0.70 0.57 0.18 0.09 0.66 0.35 0.18 0.26
polyethylene tibia, 22.8 22.7 20.9 0
(1.81-3.09) (0.12-0.59) (0.01-0.31) (0.43-1.15) (0.33-0.98) (0.07-0.47) (0.02-0.35) (0.40-1.09) (0.18-0.70) (0.07-0.47) (0.12-0.59)
with patella
monobloc
3.16 0.48 0.10 0.81 0.78 0.21 0.06 0.45 0.26 0.33 0.25 0.21
polyethylene tibia, 81.9 81.4 71.9
(2.80-3.57) (0.35-0.65) (0.05-0.20) (0.63-1.03) (0.61-1.00) (0.13-0.33) (0.03-0.15) (0.33-0.62) (0.17-0.39) (0.23-0.48) (0.16-0.38) (0.13-0.35)
without patella
pre-assembled/
13.60 0.50 1.51 7.05 1.51 1.01 0.50 0.50 1.01 1.51 0.51 1.45
hinged/linked, with 2.0 2.0 1.4
(9.33-19.84) (0.07-3.58) (0.49-4.69) (4.18-11.91) (0.49-4.69) (0.25-4.03) (0.07-3.58) (0.07-3.58) (0.25-4.03) (0.49-4.69) (0.07-3.60) (0.36-5.79)
patella
¹The indication implant failure, as reported in annual reports up to 2013, has been renamed implant wear as this reflects the wearing down of the implant but distinguishes from the implant itself breaking.
2
Other indication now includes other indications not listed, implant fracture and incorrect sizing.
3
Stiffness was asked in versions MDSv2, v3, v6 and v7 of the clinical assessment forms for joint replacement/revision surgery and hence, for these reasons, there are fewer prosthesis-years at risk.
4
Progressive arthritis was asked in versions MDSv3, v6 and v7 of the clinical assessment forms for joint replacement/revision surgery and hence, for these reasons, there are fewer prosthesis-years at risk.
187
188

Prosthe- Revisions
pre-assembled/
10.29 0.36 0.84 3.83 2.15 0.84 0.60 0.48 0.84 0.96 0.37 0.79
hinged/linked, 8.4 8.1 6.4
(8.33-12.71) (0.12-1.11) (0.40-1.76) (2.71-5.42) (1.36-3.42) (0.40-1.76) (0.25-1.44) (0.18-1.28) (0.40-1.76) (0.48-1.91) (0.12-1.14) (0.33-1.89)
without patella
4.35 0.82 0.16 0.60 1.51 0.15 0.32 0.73 0.38 0.57 0.36 0.42
All uncemented 386.8 369.4 230.3
(4.14-4.56) (0.73-0.91) (0.12-0.20) (0.53-0.69) (1.39-1.64) (0.12-0.20) (0.27-0.39) (0.65-0.82) (0.32-0.45) (0.50-0.65) (0.30-0.42) (0.34-0.51)
18.4 18.2 13.9
with patella (3.63-5.58) (0.15-0.72) (0.08-0.58) (0.45-1.28) (1.27-2.52) (0.15-0.72) (0.08-0.58) (0.79-1.81) (0.41-1.21) (0.05-0.50) (0.08-0.59) (0.01-0.51)
130.1 123.2 75.7
without patella (4.10-4.83) (0.65-0.95) (0.04-0.14) (0.47-0.74) (1.50-1.95) (0.10-0.24) (0.24-0.44) (0.54-0.83) (0.28-0.49) (0.49-0.77) (0.26-0.47) (0.37-0.69)
unconstrained, 5.78 0.89 0.22 0.89 2.22 0.11 1.22 1.78 1.00 0.78 0.61
9.0 8.1 4.1 0
unconstrained,
3.98 0.81 0.17 0.56 1.28 0.11 0.26 0.64 0.29 0.50 0.33 0.45
mobile, without 200.2 192.2 119.1
(3.71-4.27) (0.69-0.94) (0.13-0.24) (0.46-0.67) (1.13-1.45) (0.08-0.17) (0.20-0.35) (0.54-0.76) (0.22-0.37) (0.42-0.61) (0.26-0.42) (0.35-0.59)
patella
5.3 4.9 2.8 0
fixed, with patella (6.05-11.00) (0.63-2.79) (0.28-2.02) (0.63-2.79) (1.86-4.96) (0.39-2.28) (0.18-1.77) (0.89-3.28) (0.28-2.02) (0.18-1.77) (0.43-2.46)
21.8 20.8 12.9
fixed, without patella (4.56-6.53) (0.85-1.81) (0.10-0.55) (0.35-1.03) (1.15-2.23) (0.07-0.49) (0.31-0.97) (0.63-1.48) (0.41-1.14) (0.74-1.64) (0.26-0.90) (0.08-0.72)
other constraints,
0.2 0 0 0 0 0 0 0 0 0 0 0.2 0 0.1 0
with patella
other constraints, 4.17 2.60 0.52 1.56 0.52 0.52 0.52 1.06
1.9 0 0 0 1.9 1.8 0
without patella (2.08-8.33) (1.08-6.26) (0.07-3.70) (0.50-4.84) (0.07-3.70) (0.07-3.70) (0.07-3.70) (0.26-4.22)
3.54 0.58 0.15 0.83 1.09 0.13 0.31 0.58 0.31 0.27 0.20 0.28
All hybrid 89.0 81.5 42.5
(3.17-3.95) (0.45-0.77) (0.08-0.25) (0.66-1.04) (0.89-1.33) (0.08-0.24) (0.22-0.46) (0.45-0.77) (0.22-0.46) (0.18-0.40) (0.12-0.32) (0.16-0.50)
unconstrained, fixed, 2.63 0.43 0.13 0.69 0.86 0.17 0.39 0.65 0.04 0.13 0.09
23.2 21.1 7.9 0
with patella (2.05-3.38) (0.23-0.80) (0.04-0.40) (0.42-1.13) (0.56-1.34) (0.06-0.46) (0.20-0.75) (0.39-1.07) (0.01-0.31) (0.04-0.40) (0.02-0.38)
41.6 37.2 19.9
without patella (2.92-4.05) (0.41-0.89) (0.05-0.29) (0.60-1.17) (0.73-1.34) (0.04-0.26) (0.16-0.51) (0.24-0.63) (0.24-0.63) (0.20-0.57) (0.09-0.39) (0.10-0.61)
unconstrained, 5.22 0.43 0.43 2.61 0.43 0.43 0.94
2.3 0 0 0 0 2.1 1.8 0
mobile, with patella (2.96-9.19) (0.06-3.09) (0.06-3.09) (1.17-5.81) (0.06-3.09) (0.06-3.09) (0.24-3.76)
unconstrained,
4.06 0.56 0.24 0.80 1.51 0.48 0.88 0.72 0.48 0.17 0.46
mobile, without 12.6 0 11.9 8.7
(3.09-5.35) (0.27-1.17) (0.08-0.74) (0.43-1.48) (0.97-2.37) (0.21-1.06) (0.49-1.58) (0.37-1.38) (0.21-1.06) (0.04-0.67) (0.17-1.22)
patella
2
3
4

Prosthe- Revisions
posterior-stabilised, 6.92 1.98 1.98 2.97 0.99 0.49 0.52
2.0 0 0 0 0 1.9 1.2 0
fixed, with patella (4.10-11.68) (0.74-5.27) (0.74-5.27) (1.33-6.60) (0.25-3.95) (0.07-3.51) (0.07-3.71)
posterior-stabilised, 4.47 0.30 1.49 1.19 1.19 0.58
3.4 0 0 0 0 0 3.2 0 1.7
fixed, without patella (2.70-7.42) (0.04-2.12) (0.62-3.58) (0.45-3.18) (0.45-3.18) (0.08-4.10)
other constraints, 4.00 1.60 0.80 0.40 0.40 0.80 0.40 0.40
2.5 0 0 0 2.5 0.4 0
with patella (2.15-7.44) (0.60-4.27) (0.20-3.20) (0.06-2.84) (0.06-2.84) (0.20-3.20) (0.06-2.84) (0.06-2.84)
other constraints, 5.88 0.65 0.65 0.65 0.65 1.96 0.65 0.67 2.30
1.5 0 0 0 1.5 0.9
without patella (3.06-11.29) (0.09-4.63) (0.09-4.63) (0.09-4.63) (0.09-4.63) (0.63-6.07) (0.09-4.63) (0.09-4.75) (0.57-9.18)
All unicondylar, 11.33 2.12 0.61 0.46 3.14 0.20 1.12 0.91 0.54 1.50 0.17 3.46
670.4 649.6 465.7
cemented (11.08-11.59) (2.02-2.24) (0.56-0.68) (0.41-0.52) (3.01-3.28) (0.17-0.24) (1.04-1.20) (0.84-0.98) (0.49-0.60) (1.41-1.59) (0.14-0.21) (3.29-3.63)
8.14 1.56 0.09 0.54 2.22 0.19 0.72 0.53 0.39 0.94 0.15 2.72
fixed 194.1 191.8 169.9
(7.75-8.55) (1.39-1.75) (0.06-0.15) (0.44-0.65) (2.02-2.44) (0.14-0.26) (0.61-0.85) (0.43-0.64) (0.31-0.48) (0.82-1.09) (0.11-0.22) (2.49-2.98)
12.72 2.29 0.91 0.43 3.46 0.19 1.29 1.08 0.61 1.81 0.17 3.90
mobile 424.4 407.3 266.1
(12.39-13.07) (2.15-2.44) (0.82-1.00) (0.37-0.50) (3.29-3.64) (0.15-0.23) (1.19-1.41) (0.99-1.19) (0.54-0.69) (1.69-1.94) (0.14-0.22) (3.67-4.14)
monobloc 11.85 2.89 0.15 0.46 4.01 0.37 1.19 0.87 0.60 1.02 0.24 3.68
51.9 50.5 29.6
polyethylene tibia (10.95-12.82) (2.46-3.39) (0.08-0.31) (0.31-0.69) (3.50-4.59) (0.23-0.57) (0.93-1.53) (0.65-1.16) (0.42-0.85) (0.78-1.34) (0.14-0.42) (3.05-4.44)
All unicondylar, 8.53 0.92 1.30 0.49 1.45 0.61 0.84 0.76 0.42 1.22 0.12 2.12
120.3 120.1 115.6
uncemented/hybrid (8.03-9.07) (0.77-1.11) (1.12-1.53) (0.38-0.63) (1.25-1.68) (0.49-0.77) (0.69-1.02) (0.62-0.94) (0.32-0.56) (1.04-1.44) (0.07-0.20) (1.87-2.40)
10.81 1.67 0.15 5.02 0.15 1.37 1.07 0.46 1.98 0.31 2.64
fixed 6.6 0 6.4 5.3
(8.56-13.64) (0.93-3.02) (0.02-1.08) (3.57-7.07) (0.02-1.08) (0.71-2.63) (0.51-2.24) (0.15-1.42) (1.15-3.41) (0.08-1.24) (1.56-4.46)
8.43 0.83 1.41 0.53 1.22 0.66 0.79 0.76 0.44 1.20 0.09 2.09
mobile 110.3 110.2 107.2
(7.91-8.99) (0.68-1.02) (1.21-1.65) (0.41-0.69) (1.03-1.45) (0.53-0.83) (0.64-0.97) (0.61-0.94) (0.33-0.58) (1.01-1.42) (0.05-0.17) (1.83-2.38)
monobloc 7.54 2.32 1.74 1.45 0.29 0.58 0.58 2.21
3.4 0 0 0 0 3.4 3.2
polyethylene tibia (5.14-11.08) (1.16-4.64) (0.78-3.88) (0.60-3.49) (0.04-2.06) (0.15-2.32) (0.15-2.32) (1.05-4.63)
19.63 4.23 0.64 0.43 2.32 0.16 1.64 0.91 1.15 2.99 0.43 9.62
Patellofemoral 106.2 103.7 78.9
(18.81-20.50) (3.85-4.64) (0.50-0.81) (0.32-0.58) (2.04-2.62) (0.10-0.26) (1.41-1.90) (0.75-1.11) (0.96-1.37) (2.68-3.34) (0.32-0.58) (8.96-10.32)
Multi 16.31 3.31 0.71 0.47 1.42 0.24 1.42 0.95 0.95 3.31 6.67
4.2 4.2 0 3.9
Unicompartmental (12.88-20.65) (1.96-5.59) (0.23-2.20) (0.12-1.89) (0.64-3.16) (0.03-1.68) (0.64-3.16) (0.35-2.52) (0.35-2.52) (1.96-5.59) (4.54-9.79)
Unclassified
5.50 0.77 0.19 0.86 1.60 0.18 0.45 0.80 0.35 0.76 0.30 1.06
191.4 181.7 112.5
(5.17-5.84) (0.66-0.91) (0.14-0.26) (0.74-1.00) (1.43-1.79) (0.13-0.25) (0.37-0.56) (0.68-0.94) (0.28-0.44) (0.65-0.90) (0.23-0.39) (0.88-1.27)
2
3
4
189
190
Table 3.K11 PTIR estimates of indications for revision (95% CI) by years following primary knee replacement.
Pros- Number of revisions per 1,000 prosthesis-years for: Stiffness3 Progressive arthritis4
Time thesis- Prosthe- Revisions Prosthe- Revisions
since years at Aseptic Peri- sis-years per 1,000 sis-years per 1,000
primary risk Dislocation/ loosening prosthetic Implant Malalign- Other at risk prosthe at risk prosthe
(years) (x1,000) All causes Pain Subluxation Infection / Lysis fracture wear1 Instability ment indication2 (x1,000) sis-years (x1,000) sis-years
4.53 0.68 0.16 0.87 1.21 0.17 0.29 0.64 0.32 0.50 0.29 0.73
All cases 8,921.4 8,665.1 6,503.2
(4.49-4.58) (0.66-0.70) (0.16-0.17) (0.85-0.89) (1.18-1.23) (0.16-0.18) (0.28-0.30) (0.63-0.66) (0.31-0.34) (0.48-0.51) (0.27-0.30) (0.71-0.75)
4.95 0.48 0.38 1.95 0.61 0.30 0.18 0.53 0.31 0.57 0.30 0.26
<1 1,323.7 1,303.4 1,114.6
(4.83-5.07) (0.44-0.52) (0.35-0.42) (1.87-2.02) (0.57-0.65) (0.27-0.33) (0.16-0.20) (0.49-0.57) (0.28-0.34) (0.53-0.61) (0.27-0.33) (0.24-0.30)
6.39 1.30 0.19 1.19 1.51 0.12 0.19 0.95 0.55 0.77 0.54 0.91
1 to 3 2,319.7 2,280.4 1,915.0
(6.29-6.49) (1.25-1.34) (0.17-0.21) (1.14-1.23) (1.46-1.56) (0.11-0.13) (0.18-0.21) (0.92-1.00) (0.52-0.58) (0.73-0.80) (0.51-0.57) (0.87-0.96)
4.06 0.73 0.10 0.61 1.26 0.12 0.20 0.60 0.32 0.44 0.26 0.72
3 to 5 1,820.1 1,783.0 1,437.1
(3.97-4.16) (0.69-0.77) (0.08-0.11) (0.58-0.65) (1.21-1.31) (0.10-0.13) (0.18-0.22) (0.57-0.64) (0.29-0.34) (0.41-0.47) (0.24-0.29) (0.68-0.77)
3.37 0.43 0.09 0.48 1.14 0.13 0.26 0.48 0.22 0.36 0.15 0.77
5 to 7 1,352.0 1,317.6 994.3
(3.27-3.47) (0.39-0.46) (0.08-0.11) (0.45-0.52) (1.09-1.20) (0.11-0.15) (0.24-0.29) (0.44-0.52) (0.20-0.25) (0.32-0.39) (0.13-0.17) (0.72-0.83)
3.23 0.28 0.10 0.34 1.16 0.18 0.42 0.47 0.17 0.30 0.10 0.89
7 to 10 1,299.4 1,253.3 819.2
(3.13-3.33) (0.25-0.31) (0.08-0.12) (0.31-0.37) (1.10-1.22) (0.16-0.21) (0.39-0.46) (0.44-0.51) (0.14-0.19) (0.27-0.33) (0.09-0.12) (0.83-0.96)
3.67 0.18 0.13 0.35 1.37 0.25 0.75 0.58 0.15 0.29 0.08 0.75
10 to 13 616.7 578.2 220.1
(3.52-3.82) (0.15-0.22) (0.10-0.16) (0.31-0.40) (1.28-1.46) (0.21-0.29) (0.69-0.83) (0.52-0.64) (0.12-0.18) (0.25-0.33) (0.06-0.11) (0.64-0.87)
3.50 0.14 0.10 0.28 1.44 0.26 0.86 0.51 0.11 0.21 0.10 0.43
13 to 15 148.2 126.9 2.3
(3.21-3.82) (0.09-0.22) (0.06-0.17) (0.20-0.38) (1.26-1.65) (0.19-0.35) (0.73-1.03) (0.41-0.64) (0.07-0.18) (0.15-0.30) (0.06-0.18) (0.06-3.05)
3.56 0.28 0.15 0.28 1.51 0.28 1.10 0.58 0.08 0.30
15 to 17 39.8 22.4 0 0.5 0
(3.02-4.20) (0.15-0.50) (0.07-0.34) (0.15-0.50) (1.17-1.94) (0.15-0.50) (0.82-1.48) (0.38-0.87) (0.02-0.23) (0.17-0.53)
2.74 1.64 0.55 0.55 0.55
≥17 1.8 0 0 0 0 0 0.0 0 0.0 0
(1.14-6.58) (0.53-5.10) (0.08-3.89) (0.08-3.89) (0.08-3.89)
1
The indication implant failure, as reported in annual reports up to 2013, has been renamed implant wear as this reflects the wearing down of the implant but distinguishes from the implant itself breaking.
2
3
4
3.3.5 Mortality after primary probability of death (see survival analysis methods
note in section 3.1). Of the 1,357,077 records of a
knee surgery primary knee replacement, 22,272 unknown knee
In this section we describe the mortality of the cohort type records were excluded and there were 13,510
up to 15 years from primary operation, according bilateral operations in which the patient had both knees
to gender and age group. Deaths recorded after 31 replaced on the same day; here the second of the two
December 2020 have not been included in the analysis. has been excluded, leaving 1,184,306 TKR procedures
For simplicity, we have not taken into account whether (of whom 226,710 had died before the end of 2020)
the patient had a first (or further) joint revision after the and 137,399 UKR procedures (of whom 12,895 died
primary operation when calculating the cumulative before the end of 2020).
Table 3.K12 (a) KM estimates of cumulative mortality (95% CI) by age and gender, in primary TKR.
Age group Time since primary

(years) N 30 days 90 days 1 year 5 years 10 years 15 years
All primary 0.16 0.30 1.03 8.71 26.08 48.11
1,184,306
TKR cases (0.15-0.17) (0.29-0.31) (1.01-1.05) (8.65-8.77) (25.96-26.19) (47.89-48.33)
Males
0.04 0.08 0.29 2.09 6.10 11.99
<55 28,934
(0.02-0.07) (0.05-0.12) (0.23-0.36) (1.91-2.28) (5.72-6.50) (11.11-12.93)
0.05 0.10 0.36 2.92 8.74 17.77
55 to 59 42,028
(0.03-0.08) (0.07-0.14) (0.31-0.42) (2.75-3.11) (8.36-9.13) (16.95-18.62)
0.07 0.13 0.47 4.06 11.74 25.53
60 to 64 76,325
(0.06-0.10) (0.11-0.16) (0.43-0.52) (3.91-4.22) (11.43-12.06) (24.80-26.27)
0.10 0.18 0.67 5.84 17.78 37.61
65 to 69 99,210
(0.08-0.12) (0.15-0.21) (0.62-0.73) (5.68-6.00) (17.44-18.12) (36.88-38.36)
0.14 0.27 1.05 9.29 28.46 56.57
70 to 74 105,714
(0.12-0.16) (0.24-0.30) (0.99-1.11) (9.10-9.49) (28.07-28.86) (55.82-57.32)
0.29 0.52 1.79 15.03 44.70 76.74
75 to 79 85,513
(0.25-0.32) (0.47-0.57) (1.70-1.88) (14.76-15.30) (44.22-45.18) (76.02-77.46)

0.58 0.99 3.02 24.04 63.97 91.38
80 to 84 47,792
(0.51-0.65) (0.90-1.08) (2.87-3.18) (23.61-24.48) (63.32-64.61) (90.65-92.07)
1.10 1.93 5.71 38.80 82.53 97.30
≥85 18,358
(0.95-1.26) (1.75-2.15) (5.38-6.05) (37.99-39.61) (81.64-83.39) (96.40-98.03)
Females
0.03 0.06 0.21 1.61 4.58 9.50
<55 41,309
(0.01-0.05) (0.04-0.08) (0.17-0.26) (1.48-1.75) (4.30-4.88) (8.80-10.24)
0.03 0.06 0.26 2.08 6.29 14.11
55 to 59 56,179
(0.02-0.05) (0.04-0.08) (0.22-0.30) (1.95-2.22) (6.01-6.57) (13.44-14.81)
0.03 0.08 0.31 2.74 8.70 19.48
60 to 64 91,581
(0.02-0.05) (0.07-0.11) (0.27-0.35) (2.62-2.86) (8.45-8.96) (18.86-20.12)
0.07 0.12 0.43 3.90 12.81 29.85
65 to 69 123,453
(0.05-0.08) (0.10-0.14) (0.39-0.47) (3.78-4.02) (12.54-13.08) (29.22-30.49)
0.09 0.18 0.63 6.00 20.66 46.23
70 to 74 138,945
(0.08-0.11) (0.16-0.20) (0.59-0.68) (5.86-6.14) (20.35-20.97) (45.57-46.90)
0.16 0.30 1.12 10.20 33.99 66.72
75 to 79 122,809
(0.14-0.18) (0.27-0.34) (1.06-1.18) (10.01-10.39) (33.62-34.37) (66.06-67.37)
0.27 0.55 1.88 16.43 51.83 84.72
80 to 84 75,160
(0.24-0.31) (0.50-0.60) (1.78-1.98) (16.13-16.73) (51.31-52.35) (84.05-85.38)
0.58 1.19 3.49 28.67 73.36 95.14
≥85 30,996
(0.50-0.67) (1.08-1.32) (3.29-3.70) (28.10-29.24) (72.62-74.10) (94.45-95.78)
Note: Excludes 8,819 bilateral operations performed on the same day.
Table 3.K12 (a) on page 191 shows Kaplan-Meier females following UKR than TKR, but these figures do
estimates of cumulative percentage mortality at 30 not adjust for selection and hence do not account for
days, 90 days and at 1, 5, 10 and 15 years from the residual confounding (Hunt et al., 2018).
primary knee replacement, for all cases and by age
and gender. Fewer men than women have had a Note: These cases were not censored when further
primary knee replacement and, proportionally, more revision surgery was undertaken. While such surgery
women than men undergo surgery above the age of may have contributed to the overall mortality, the impact
75. Males, particularly in the older age groups, had of this is not investigated in this report. Furthermore,
a higher cumulative percentage probability of dying exclusions for unknown knee type and same-day
in the short or longer term after their primary knee bilateral operations were not mutually exclusive; there
replacement operation than females in the equivalent was an overlap of 410 cases of unknown knee types
age group. The mortality rates are lower in males and with same day bilateral procedures.
Table 3.K12 (b) KM estimates of cumulative mortality (95% CI) by age and gender, in primary unicompartmental
replacements. Blue italics signify that fewer than 250 cases remained at risk at these time points.
Time since primary

Age group (years) N 30 days 90 days 1 year 5 years 10 years 15 years
0.03 0.08 0.39 4.16 13.31 27.45
All unicondylar 121,986
(0.03-0.05) (0.06-0.09) (0.36-0.43) (4.04-4.30) (13.04-13.60) (26.85-28.07)
Males
0.01 0.03 0.18 1.23 3.65 8.30
<55 10,128
(0.00-0.07) (0.01-0.09) (0.11-0.29) (1.02-1.50) (3.14-4.23) (6.91-9.97)
0.03 0.04 0.20 1.71 5.95 12.42
55 to 59 10,182
(0.01-0.09) (0.01-0.11) (0.13-0.31) (1.44-2.03) (5.30-6.68) (11.02-13.99)
0.05 0.09 0.34 2.94 8.69 20.47
60 to 64 12,844
(0.03-0.11) (0.05-0.16) (0.26-0.46) (2.62-3.29) (8.03-9.40) (18.81-22.26)
0.01 0.05 0.34 4.34 14.54 29.69
65 to 69 12,426
(0.00-0.06) (0.02-0.11) (0.25-0.46) (3.94-4.77) (13.65-15.50) (27.74-31.75)
0.02 0.07 0.59 7.26 22.57 48.23
70 to 74 9,822
(0.01-0.08) (0.03-0.15) (0.46-0.77) (6.66-7.90) (21.32-23.89) (45.58-50.96)
0.05 0.17 1.00 11.28 37.85 71.13
75 to 79 6,100
(0.02-0.15) (0.09-0.31) (0.78-1.29) (10.36-12.26) (35.97-39.78) (67.94-74.25)
0.11 0.25 1.81 20.02 53.90 85.85
80 to 84 2,801
(0.03-0.33) (0.12-0.53) (1.37-2.38) (18.27-21.90) (50.99-56.86) (82.09-89.18)
0.55 0.78 3.56 33.80 80.10 97.55
≥85 905
(0.23-1.33) (0.37-1.62) (2.52-5.03) (30.15-37.76) (75.32-84.48) (90.97-99.67)
Hunt LP, Whitehouse MR, Howard PW, Ben-Shlomo Y, Blom AW. Using long term mortality to determine which peri-operative risk factors of mortality following hip
and knee replacement may be causal. Sci Rep. 2018 Oct 9;8(1):15026.
Time since primary

Age group (years) N 30 days 90 days 1 year 5 years 10 years 15 years
Females
0.02 0.03 0.06 0.78 2.64 4.71
<55 11,464
(0.00-0.07) (0.01-0.08) (0.03-0.13) (0.62-0.99) (2.24-3.11) (3.89-5.71)
0.01 0.01 0.07 1.02 3.85 7.90
55 to 59 9,267
(0.00-0.08) (0.00-0.08) (0.03-0.15) (0.80-1.29) (3.32-4.46) (6.78-9.19)
0.01 0.01 0.13 1.76 5.84 13.91

60 to 64 9,846
(0.00-0.07) (0.00-0.07) (0.07-0.22) (1.48-2.08) (5.22-6.53) (12.36-15.63)
0.03 0.08 0.26 2.52 8.46 21.18
65 to 69 9,524
(0.01-0.10) (0.04-0.17) (0.17-0.38) (2.19-2.91) (7.68-9.33) (19.14-23.41)
0.05 0.09 0.34 3.96 13.95 34.39
70 to 74 8,122
(0.02-0.13) (0.04-0.18) (0.23-0.50) (3.48-4.50) (12.83-15.17) (31.90-37.02)
0.06 0.34 6.44 24.59 54.23
75 to 79 5,201 0
(0.02-0.18) (0.21-0.54) (5.70-7.28) (22.89-26.39) (50.92-57.61)
0.12 0.33 1.12 12.12 42.46 75.24
80 to 84 2,482
(0.04-0.37) (0.16-0.65) (0.77-1.62) (10.69-13.73) (39.64-45.40) (70.93-79.34)
0.34 0.93 2.95 20.13 61.67 97.60
≥85 872
(0.11-1.07) (0.46-1.84) (2.01-4.34) (17.13-23.57) (56.41-66.95) (90.10-99.76)
0.04 0.13 0.38 3.71 11.67 23.37
All patellofemoral 14,852
(0.02-0.09) (0.08-0.20) (0.29-0.49) (3.39-4.06) (10.99-12.39) (21.80-25.02)
0.36 2.57 7.88 25.94
All multicompartmental 561 0 0
(0.09-1.43) (1.50-4.39) (5.53-11.17) (11.44-52.40)
3.3.6 Overview of knee revisions for the DAIR procedures in MDSv7, it may be possible
to report these as distinct categories in future reports.
In this section we look at all recorded knee revision Although not all patients who undergo stage one of
procedures performed since the registry began on a two-stage revision will undergo a stage two of two-
1 April 2003 up to the end of December 2020, for all stage revision. In some cases, stage one revisions have
patients with valid patient identifiers (i.e. whose data been entered without stage two, and vice versa, making
could therefore be linked). identification of entire patient revision episodes difficult.
We have attempted to address this later in this section.
In total there were 87,535 revisions recorded on
72,493 individual patient-sides (68,973 actual patients). The NJR asks surgeons and those responsible for
In addition to the 40,451 revised primaries described healthcare delivery to ensure that when primary and
previously in this section, there were 32,042 additional revision joint replacement procedures of the hip,
revisions for a patient-side for which there is no knee, ankle, elbow or shoulder are performed, that
associated primary operation recorded in the registry. the relevant MDS form is completed and data entered
into the registry. This is a requirement mandated by
We have classified revisions as single-stage, stage the Department of Health and Social Care. For the
one of two-stage, or stage two of two-stage revisions. purposes of the annual report, revision procedures
Information on stage one and stage two of two-stage include any addition, removal or modification of the
revisions are entered into the registry separately. implants and procedures such as debridement and
Debridement and Implant Retention (DAIR) with or implant retention with or without implant exchange,
without modular exchange are included as single-stage excision arthroplasty, amputation and conversion
procedures. With the introduction of distinct indicators
to arthrodesis. For the avoidance of confusion, Table 3.K13 below gives an overview of all knee revision
completing a revision MDS form is also mandatory procedures carried out each year since April 2003.
for a procedure involving modification of a joint by There were a maximum number of 14 documented
adding another implant to another part of the joint. revision procedures associated with any individual
For the analyses of surgeon performance, hospital patient-side. The increase in the number of operations
performance and implant performance, debridement over time, until 2020 when rates were impacted by
and implant retention without implant exchange is COVID-19, reflects the increasing number of at-risk
currently excluded. implants prevailing in the dataset.
Table 3.K13 Number and percentage of revisions by procedure type and year.

Year of revision Single-stage Stage one of Stage two of Total revision joint
surgery N(%) two-stage N(%) two-stage N(%) operations
2003* 7 (1.1) <4 (0.2) 625 (98.7) 633
2004 702 (57.3) 78 (6.4) 445 (36.3) 1,225
2005 1,475 (73.7) 209 (10.4) 318 (15.9) 2,002
2006 1,947 (75.3) 282 (10.9) 356 (13.8) 2,585
2007 2,641 (75.1) 386 (11.0) 491 (14.0) 3,518

2008 3,324 (75.6) 474 (10.8) 596 (13.6) 4,394
2009 3,715 (76.3) 527 (10.8) 630 (12.9) 4,872
2010 4,182 (77.1) 573 (10.6) 670 (12.4) 5,425
2011 4,339 (77.4) 620 (11.1) 649 (11.6) 5,608
2012 5,009 (78.5) 631 (9.9) 740 (11.6) 6,380
2013 4,704 (78.4) 633 (10.6) 662 (11.0) 5,999
2014 5,078 (78.0) 736 (11.3) 700 (10.7) 6,514
2015 5,353 (79.1) 743 (11.0) 675 (10.0) 6,771
2016 5,562 (80.6) 698 (10.1) 643 (9.3) 6,903
2017 5,667 (80.6) 700 (10.0) 666 (9.5) 7,033
2018 5,601 (82.2) 618 (9.1) 598 (8.8) 6,817
2019 5,847 (83.5) 613 (8.8) 545 (7.8) 7,005
2020 3,067 (79.6) 426 (11.1) 358 (9.3) 3,851
Total 68,220 8,948 10,367 87,535
*Incomplete year.
Table 3.K14 (a) below shows the stated indications while instability, pain, wear and other indications
for the revision knee surgery. As more than one account for between 10% and 20% each. Of the two-
indication can be selected, the indications are not stage revision operations, infection is the main indication
mutually exclusive and therefore column percentages recorded in approximately 80% of either stage one or
do not add up to 100%. Aseptic loosening / lysis is the stage two procedures. Table 3.K14 (b) presents these
most common indication for revision, accounting for results, restricted to the last five years.
approximately 40% of single-stage revision operations,
Table 3.K14 (a) Number and percentage of knee revision by indication and procedure type.

Reason for revision N(%) (n=68,220) N(%) (n=8,948) N(%) (n=10,367)
Aseptic loosening / Lysis 26,072 (38.2) 1,565 (17.5) 1,750 (16.9)
Instability 11,850 (17.4) 353 (3.9) 506 (4.9)

Pain 10,137 (14.9) 364 (4.1) 529 (5.1)
Implant wear 9,542 (14.0) 288 (3.2) 314 (3.0)
Infection 5,180 (7.6) 7,628 (85.2) 7,719 (74.5)
Malalignment 5,018 (7.4) 113 (1.3) 175 (1.7)
3,904 (5.7) 200 (2.2) 165 (1.7)
Stiffness* n=68,220 n=8,948 n=9,461
9,078 (14.9) 62 (0.8) 89 (1.1)

Progressive arthritis* n=60,989 n=7,920 n=8,034
*These reasons were not recorded in the earliest phase of the registry; only in MDSv2 onwards for stiffness and MDSv3 onwards for progressive arthritis.
Note: The number of joints on which these two percentages are based is stated beside the percentage figure.
Note: Indications listed are not mutually exclusive.
Table 3.K14 (b) Number and percentage of knee revision by indication and procedure type in the last
five years.

Reason for revision N(%) (n=25,744) N(%) (n=3,055) N(%) (n=2,810)
Aseptic loosening / Lysis 8,467 (32.9) 412 (13.5) 296 (10.5)
Progressive arthritis 5,177 (20.1) 27 (0.9) 53 (1.9)

Instability 4,426 (17.2) 94 (3.1) 73 (2.6)
Implant wear 3,353 (13.0) 69 (2.3) 41 (1.5)
Infection 2,857 (11.1) 2,715 (88.9) 2,347 (83.5)
Pain 2,412 (9.4) 48 (1.6) 40 (1.4)
Malalignment 1,580 (6.1) 28 (0.9) 28 (1.0)
Stiffness 1,435 (5.6) 43 (1.4) 40 (1.4)
Dislocation / Subluxation 961 (3.7) 49 (1.6) 27 (1.0)
Note: Indications listed are not mutually exclusive.
3.3.7 Rates of knee re-revision distinct revision episodes for any patient-side was
determined to be 14). In cases where a stage one
In most instances (86%), the first revision procedure of two procedure was followed by a stage two of
was a single-stage revision, in the remaining two procedure within 365 days, we have treated this
14% it was part of a two-stage procedure. For a as a single distinct episode. This definition allows
given patient-side, the survival following the first multiple stage one procedures to occur before a new
documented revision procedure linked to a primary revision episode is triggered. In situations where the
in the registry (n=40,451) has been analysed. This first stage one procedure is not followed by a stage
analysis is restricted to patients with a linked primary two procedure within a 365-day period, the next
procedure so that there is confidence that the next occurrence of a stage one procedure was considered
observed procedure on the same joint is the first as a new revision episode.
revision episode. If there is no linked primary record
in the dataset, it cannot be determined if the first Kaplan-Meier estimates of the cumulative percentage
observed revision is the first revision or has been probability of having a subsequent revision (re-revision)
preceded by other revision episodes. The time from were calculated. There were 4,501 re-revisions and for
the first documented revision procedure (of any type) 5,169 cases the patient died without having been re-
to the time at which a second revision procedure was revised. The censoring date for the remainder was the
undertaken has been determined. For this purpose, end of 2020.
an initial stage one followed by either a stage one
Figure 3.K6 (a) plots Kaplan-Meier estimates of the
or a stage two of a two-stage procedure have been
cumulative probability of a subsequent revision in
considered to be the same revision episode and these
linked revised primary knee replacements as between
were disregarded, looking instead for the start of a
1 and 17 years since the primary operation.
second revision episode. (The maximum number of
Figure 3.K6 (a) KM estimates of cumulative re-revision, in linked primary knee replacements (shaded
area indicates point-wise 95% CI). Blue italics in the numbers at risk table signify that fewer than 250 cases
25
20

15
10
0
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
Numbers at risk
40,451 30,523 21,316 14,039 8,666 4,502 1,747 423 33
Figure 3.K6 (b) shows estimates of re-revision by type Revised uncemented / hybrid unicondylar knee
of primary knee replacement. Revised patellofemoral replacements appear to have a higher risk of re-
knee replacements have the lowest risk of re-revision revision than their cemented counterparts and are
until ten years, after which the numbers at risk fall equivalent to the rates seen for revised cemented total
below 250 and should be interpreted with caution. knee replacements until five years, after which the
Revised cemented unicondylar knee replacements numbers in the revised uncemented unicondylar group
have the next lowest risk of re-revision until 14 years become small.
when again, the numbers at risk become small.
Figure 3.K6 (b) KM estimates of cumulative re-revision by primary fixation, in linked primary knee
these time points.
25
20
15
10
0
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
Cemented 26,625 19,587 13,305 8,473 5,007 2,543 971 208 12
Uncemented 1,681 1,387 1,064 784 543 298 123 32 <4
Hybrid 315 253 206 159 118 71 27 7
Unicondylar, cemented 7,596 6,124 4,531 3,186 2,087 1,148 483 143 15
Unicondylar, uncemented/hybrid 1,027 638 355 193 121 55 17 6
Patellofemoral 2,086 1,659 1,200 777 484 229 64 14 <4
Figure 3.K6 (c) KM estimates of cumulative re-revision by years to first revision, in linked primary knee
these time points.
30
25

20
15
10
0
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
First rev. <1y 6,547 5,634 4,721 4,069 3,470 2,954 2,487 2,086 1,706 1,355 1,081 784 529 316 177 86 27 <4
First rev. 1 to 3y 14,826 13,722 12,010 10,533 9,128 7,777 6,515 5,371 4,389 3,359 2,514 1,667 1,012 538 243 74 6
First rev. 3 to 5y 7,396 6,739 5,794 4,929 4,127 3,401 2,813 2,241 1,716 1,147 725 408 195 57 <4
First rev. ≥5y 11,682 10,100 7,998 6,183 4,591 3,284 2,224 1,413 855 434 182 65 11
Figure 3.K6 (c) shows the relationship between time an 8.3% re-revision rate at one year following the first
to first revision and risk of subsequent revision. The revision, rising to 19.6% by five years; if a primary knee
earlier the primary knee replacement fails, the higher replacement is not revised until five years or more after
the risk of second revision. For example, if a primary the primary procedure, the re-revision rate is 2.2% at
knee replacement is revised within the first year of one year following the first revision, rising to 7.2% by
the primary replacement being performed, there is five years.
For those with documented primary knee replacements patellofemoral knee replacements, those who had
within the registry, Figures 3.K7 (a) to (f) show their first revision within one year of the initial primary
cumulative re-revision rates following the first knee replacement experienced the worst re-revision
revision, according to the main type of primary knee rates. However, for hybrid TKRs, the worst re-revision
replacement. We have further sub-divided each rates were experienced by those who had their first
sub-group according to the time interval from the revision within three to five years of the initial primary
primary to the first revision, i.e. less than 1 year, 1 to knee replacement. However, the numbers at risk were
3, 3 to 5 and greater than or equal to 5 years. For small in the hybrid group and therefore we advise that
cemented TKRs, uncemented TKRs, unicondylar and the results should be interpreted with caution.
Figure 3.K7 (a) KM estimates of cumulative re-revision in primary cemented TKRs by years to first
revision. Blue italics in the numbers at risk table signify that fewer than 250 cases remained at risk at these
time points.
30
25
20
15
10
0
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
First rev. <1y 4,651 3,964 3,286 2,797 2,344 1,946 1,595 1,300 1,029 797 609 421 279 159 85 38 10 <4
First rev. 1 to 3y 10,419 9,544 8,242 7,095 6,011 5,041 4,119 3,311 2,622 1,972 1,467 958 579 293 121 39 <4
First rev. 3 to 5y 4,971 4,447 3,729 3,082 2,525 2,008 1,628 1,260 949 627 385 219 108 30 <4
First rev. ≥5y 6,584 5,554 4,330 3,300 2,425 1,706 1,131 690 407 204 82 30 5
Figure 3.K7 (b) KM estimates of cumulative re-revision in primary uncemented TKRs by years to first
revision. Blue italics in the numbers at risk table signify that fewer than 250 cases remained at risk at these
time points.
30
25

20
15
10
0
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
First rev. <1y 261 234 209 192 173 162 149 135 118 96 78 62 42 27 16 7 <4 0
First rev. 1 to 3y 661 623 572 540 483 426 378 331 294 237 175 117 70 40 16 4 <4
First rev. 3 to 5y 282 272 244 217 195 174 151 124 93 58 34 20 10 <4 0
First rev. ≥5y 477 426 362 291 213 149 106 60 38 19 11 <4 <4
Figure 3.K7 (c) KM estimates of cumulative re-revision in primary hybrid TKRs by years to first
revision. Blue italics in the numbers at risk table signify that fewer than 250 cases remained at risk at
these time points.
30
25
20
15
10
0
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
First rev. <1y 50 44 41 36 34 32 27 25 23 20 17 13 9 <4 <4 <4 0 0
First rev. 1 to 3y 109 103 97 94 86 85 77 66 61 52 43 33 18 11 4 <4 0
First rev. 3 to 5y 55 49 44 38 30 25 23 22 16 11 6 4 0 0 0
First rev. ≥5y 101 85 71 63 56 45 32 23 18 11 5 0 0
Figure 3.K7 (d) KM estimates of cumulative re-revision in primary patellofemoral knee replacements
by years to first revision. Blue italics in the numbers at risk table signify that fewer than 250 cases
30
25

20
15
10
0
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
First rev. <1y 161 146 138 127 114 103 87 74 61 47 38 26 15 8 4 <4 <4 0
First rev. 1 to 3y 657 633 573 509 460 391 331 277 228 173 129 70 39 18 9 <4 0
First rev. 3 to 5y 429 406 359 325 285 241 203 166 135 83 50 23 9 <4 <4
First rev. ≥5y 839 756 589 459 341 248 156 104 60 26 12 <4 <4
Figure 3.K7 (e) KM estimates of cumulative re-revision in primary cemented unicondylar knee
replacements by years to first revision. Blue italics in the numbers at risk table signify that fewer than
250 cases remained at risk at these time points.
35
30
25
20
15
10
0
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
First rev. <1y 917 817 728 659 602 549 495 436 374 314 270 212 155 103 61 31 12 0
First rev. 1 to 3y 2,297 2,192 1,999 1,837 1,679 1,481 1,303 1,122 962 770 593 421 265 155 82 24 <4
First rev. 3 to 5y 1,322 1,263 1,152 1,041 911 803 694 582 464 332 224 125 60 20 0
First rev. ≥5y 3,060 2,738 2,245 1,766 1,339 986 694 461 287 150 61 25 <4
Figure 3.K7 (f) KM estimates of cumulative re-revision in primary uncemented / hybrid unicondylar
knee replacements by years to first revision. Blue italics in the numbers at risk table signify that fewer
than 250 cases remained at risk at these time points.
30
25

20
15
10
0
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
First rev. <1y 340 285 198 147 107 73 53 44 40 30 25 16 7 <4 <4 <4 0 0
First rev. 1 to 3y 340 303 233 183 149 112 92 76 58 33 20 10 8 5 4 0 0
First rev. 3 to 5y 152 125 97 69 47 33 25 17 13 6 6 5 <4 <4 0
First rev. ≥5y 195 154 110 73 52 31 23 15 10 5 4 <4 0
Table 3.K15 (a) KM estimates of cumulative re-revision (95% CI). Blue italics signify that fewer than 250 cases
remained at risk at these time points

Number of
Time since first revision
first revised
joints at risk
of re-revision 1 year 3 years 5 years 10 years 15 years 17 years
Primary recorded 3.56 8.97 11.92 15.89 19.51 20.01
40,451
in the NJR (3.38-3.75) (8.68-9.28) (11.56-12.28) (15.40-16.39) (18.23-20.86) (18.43-21.71)
Note: The number at risk for the 17 year estimate is only 2.
Table 3.K15 (a) shows the re-revision rate of the Table 3.K15 (b) shows that primary knee replacements
40,451 revised primary knee replacements (39,399 that fail within the first year after surgery have
(97.4%) with known knee type at primary procedure) approximately two to four times the chance of needing
that are registered in the registry. Of these, 4,501 were re-revision at each time point compared with primaries
re-revised. that last more than five years.
Table 3.K15 (b) KM estimates of cumulative re-revision (95% CI) by years since first revision.
Primary in the Number of
NJR where the first revised
first revision joints at risk of

took place: re-revision 1 year 3 years 5 years 10 years 15 years
<1 year after 8.33 16.56 19.60 23.39 27.01
6,547
primary (7.68-9.04) (15.64-17.54) (18.58-20.67) (22.16-24.67) (24.75-29.44)
1 to 3 years after 3.04 9.32 12.49 16.52 19.90
14,826
primary (2.77-3.33) (8.84-9.82) (11.92-13.09) (15.78-17.29) (18.59-21.29)
3 to 5 years after 2.50 7.04 10.19 14.74
7,396
primary (2.16-2.89) (6.44-7.69) (9.43-11.01) (13.61-15.95)
≥5 years after 2.18 5.04 7.21 9.74
11,682
primary* (1.93-2.47) (4.61-5.50) (6.64-7.83) (8.79-10.79)
*The maximum of this interval was 17.5 years.

Note: Data not presented for 17 years due to low numbers.
Table 3.K15 (c) shows cumulative re-revision rates at rates were demonstrated in those where the initial
1, 3, 5, 10 and 15 years following the first revision for primary had been a cemented TKR, hybrid TKR or
those with documented primary knee replacements an uncemented unicondylar although the confidence
within the registry, broken down by type of knee intervals broadly overlap after five years in the cemented
replacement, constraint, mobility and whether a patella TKR group and earlier in the other groups.
component was recorded. Overall, the worst re-revision
Table 3.K15 (c) KM estimates of cumulative re-revision (95% CI) by fixation and constraint and whether a patella
component was recorded. Blue italics signify that fewer than 250 cases remained at risk at these time points.

Knee type Constraint N 1 year 3 years 5 years 10 years 15 years
3.56 8.97 11.92 15.89 19.51
All types 40,451
(3.38-3.75) (8.68-9.28) (11.56-12.28) (15.40-16.39) (18.23-20.86)
3.29 9.43 12.37 14.48
Unclassified 1,052
(2.36-4.58) (7.73-11.47) (10.35-14.75) (12.06-17.33)
4.04 9.87 12.84 16.91 20.20
Cemented 26,625
(3.81-4.29) (9.49-10.26) (12.38-13.30) (16.28-17.57) (18.89-21.59)
unconstrained, fixed, 5.02 11.21 13.92 17.57 19.74
4,893
with patella (4.43-5.68) (10.30-12.21) (12.85-15.08) (16.13-19.11) (17.60-22.10)
unconstrained, fixed, 3.52 9.25 12.40 15.97 18.78
11,525
without patella (3.19-3.87) (8.69-9.84) (11.72-13.11) (15.05-16.95) (17.01-20.71)
unconstrained, mobile, 3.51 9.46 12.73 19.28
998

without patella (2.52-4.88) (7.73-11.54) (10.65-15.17) (16.30-22.73)
posterior-stabilised, 4.78 10.36 13.50 17.20
3,260
fixed, with patella (4.08-5.59) (9.28-11.55) (12.19-14.93) (15.41-19.16)
posterior-stabilised, 3.51 9.13 11.65 16.14 18.43
4,381
fixed, without patella (3.00-4.11) (8.25-10.10) (10.61-12.79) (14.62-17.82) (15.64-21.65)
3.52 8.81 12.35 16.29 19.43
Uncemented 1,681
(2.73-4.52) (7.50-10.34) (10.75-14.18) (14.22-18.63) (16.29-23.08)
unconstrained, mobile, 4.11 7.89 10.56 14.97
797
without patella (2.93-5.77) (6.17-10.08) (8.50-13.09) (12.17-18.34)
4.57 8.58 12.84 17.56
Hybrid 315
(2.73-7.60) (5.87-12.45) (9.36-17.48) (13.03-23.44)
Unicondylar, 2.34 6.81 9.60 13.78 18.59
7,596
cemented (2.02-2.71) (6.24-7.44) (8.88-10.37) (12.77-14.87) (15.48-22.25)
2.29 8.05 11.07 15.83
fixed 1,580
(1.65-3.18) (6.71-9.66) (9.38-13.04) (13.44-18.61)
2.46 6.65 9.25 13.50 17.67
mobile 5,401
(2.08-2.92) (5.98-7.38) (8.43-10.15) (12.33-14.78) (14.24-21.82)
Unicondylar,
4.75 10.49 13.48 15.65
uncemented/ 1,027
(3.58-6.29) (8.55-12.82) (11.04-16.40) (12.27-19.84)
hybrid
5.02 10.62 13.94 15.16
mobile 930
(3.76-6.69) (8.59-13.10) (11.28-17.17) (12.12-18.87)
1.47 4.95 7.47 10.74
Patellofemoral 2,086
(1.03-2.09) (4.04-6.06) (6.27-8.88) (9.02-12.76)
Note: Maximum interval was 17 years.
3.3.8 Reason for knee re-revision
Table 3.K16 (a) Number of revisions by indication for all revisions.
Reason for revision All recorded revisions, N(%)

Aseptic loosening / Lysis 29,387 (33.6)
Infection 20,527 (23.5)
Instability 12,709 (14.5)
Pain 11,030 (12.6)

Implant wear 10,144 (11.6)
Malalignment 5,306 (6.1)
Periprosthetic fracture 3,358 (3.8)
Dislocation / Subluxation 3,055 (3.5)
Other indication 8,608 (9.8)
Stiffness* 4,269 (4.9)
Progressive Arthritis** 9,229 (12.0)
*Stiffness as a reason for revision was not recorded in MDSv1 and as such was only a potential reason for revision among a total of 86,629 revisions as opposed to
87,535 revisions for the other reasons.
**Progressive arthritis as a reason for revision was not recorded in MDSv1 or MDSv2 and as such was only a potential reason for revision among a total of 76,943
revisions, as opposed to 87,535 revisions for the other reasons.
Table 3.K16 (b) Number of revisions by indication for first linked revision and second linked re-revision.
First linked revision Second linked revision

Subsequently
Reason for revision N re-revised, N(%) N
Aseptic loosening / Lysis 10,759 1,057 (9.8) 1,115
Infection 7,799 1,443 (18.5) 1,750
Pain 6,052 680 (11.2) 461

Instability 5,730 602 (10.5) 781
Malalignment 2,883 262 (9.1) 247
Implant wear 2,587 243 (9.4) 183
Periprosthetic fracture 1,497 107 (7.1) 112
Dislocation / Subluxation 1,468 212 (14.4) 185
Other indication 4,421 443 (10.0) 336
Stiffness* 2,474 289 (11.7) 281
Progressive Arthritis** 4,744 230 (4.8) 115
*Stiffness as a reason for revision was not recorded in MDSv1 and as such was only a potential reason for revision among a total of 39,374 linked revisions as
opposed to 40,451 linked revisions for the other reasons.
**Progressive arthritis as a reason for revision was not recorded in MDSv1 or MDSv2 and as such was only a potential reason for revision among a total of 29,572
linked revisions, as opposed to 40,451 linked revisions for the other reasons.
Tables 3.K16 (a) and (b) show a breakdown of the subsequently re-revised. The final column reports
stated indications for the first revision and for any the indications for all the second linked revisions.
second revision (please note the indications are It is interesting to note that infection, dislocation
not mutually exclusive). Table 3.K16 (a) shows the / subluxation, instability and stiffness are more
indications for all knee revisions recorded in the common indications for second revision than for a first
registry and Table 3.K16 (b) reports the indications revision. This reflects the roles that infection, surgical
for the first linked revision and the number and complexity and soft tissue elements contribute to the
percentage of first recorded revisions that were outcome of revision knee replacement.
Table 3.K17 (a) Number of revisions by year.

Number of first revisions (%) with the
Year of first revision in the NJR* Number of first revisions associated primary recorded in the NJR
2003 625 12 (1.9)
2004 1,168 83 (7.1)
2005 1,844 280 (15.2)
2006 2,340 509 (21.8)
2007 3,141 882 (28.1)
2008 3,810 1,390 (36.5)

2009 4,193 1,832 (43.7)
2010 4,611 2,202 (47.8)
2011 4,690 2,358 (50.3)
2012 5,296 2,977 (56.2)
2013 4,912 2,845 (57.9)
2014 5,248 3,225 (61.5)
2015 5,417 3,521 (65.0)
2016 5,504 3,770 (68.5)
2017 5,607 3,982 (71.0)
2018 5,448 4,036 (74.1)
2019 5,653 4,280 (75.7)
2020 2,986 2,267 (75.9)
Total 72,493 40,451 (55.8)
Table 3.K17 (b) Number of revisions by year, stage, and whether or not primary is in the NJR.
Single-stage First documented stage of two-stage

Primary not in the Primary in the NJR Primary not in the Primary in the NJR
Year of (first) revision NJR total per year total per year NJR total per year total per year
2003 5 <4 608 10
2004 647 47 438 36
2005 1,235 202 329 78
2006 1,493 385 338 124
2007 1,861 667 398 215
2008 2,036 1,091 384 299

2009 1,984 1,504 377 328
2010 2,060 1,809 349 393
2011 2,039 1,930 293 428
2012 2,061 2,510 258 467
2013 1,827 2,411 240 434
2014 1,812 2,728 211 497
2015 1,711 3,042 185 479
2016 1,572 3,331 162 439
2017 1,484 3,520 141 462
2018 1,314 3,596 98 440
2019 1,281 3,868 92 412
2020 650 1,995 69 272
Total 27,072 34,638 4,970 5,813
Tables 3.K17 (a) and (b) show that the numbers of 1.00 (0.90-1.12)), which may reflect the fact that this
revisions and the relative proportion of revisions with patient group was younger at the time of their first
an associated primary in the registry increased with revision, with a median age of 68 (IQR 61 to 75) years,
time. Approximately 75% of those revisions performed compared to the group without primaries documented
in 2020 had a linked primary in the registry. We in the registry who had a median age of 73 (IQR 65 to
propose that this is likely to reflect improved data 79) years. The percentage of males was similar in both
capture over time, improved linkability of records and groups (45.1% versus 46.9% respectively).
the longevity of knee replacements, with a proportion
of primaries being revised having been performed 3.3.10 Conclusions
before registry data capture began or are outside the
coverage of the registry. There are now over 1.3 million primary knee
replacements recorded in the registry with a maximum
3.3.9 90-day mortality after follow-up of 17.75 years, making this the largest
dataset of its kind in the world. Of these, 96.6% of the
knee revision procedures were performed for osteoarthritis as the
The overall cumulative percentage probability of only indication. Approximately 90% of the procedures
mortality at 90 days after knee revision was lower are TKRs, 9% medial or lateral unicondylar knee
in the cases with their primaries documented in replacements and 1% patellofemoral replacements.
the registry compared with the remainder (Kaplan- These proportions have remained relatively
Meier estimates 0.72 (95% CI 0.65-0.81) versus constant over time but the proportion of unicondylar
knee replacements has risen slightly, reaching high revision rates. The difference in revision rates
approximately 10% for the first time in 2017 and rising rises from the under 55 age group up to the 65 to 74
to 11.5% in 2020. The popularity of uncemented age group, and then declines again in the over 75s.
unicondylar replacements has risen relatively rapidly.
These made up less than 1% of knee replacements in The most common causes of revision across all
2010 and now account for 4.3%, that is over a third primary knee replacements were for aseptic loosening
of the unicondylar knee replacements performed. / lysis, infection, progressive arthritis, pain and
Cemented, unconstrained (cruciate retaining), fixed instability. For uncemented TKRs, the incidence of
bearing TKR remains by far the most common type of revision for pain and aseptic loosening / lysis, wear
knee replacement, followed by cemented, posterior and ‘other’ indications was higher but the risk of
stabilised, fixed bearing TKR. Patients who received revision for infection lower than for cemented TKR.
unicondylar or patellofemoral knee replacement were For cemented unicondylar knee replacements, the
typically younger than those receiving a TKR. Both highest risk of revision was for progressive arthritis,
TKR and patellofemoral replacement are more likely to aseptic loosening / lysis and pain. For uncemented
be performed on females, whereas unicondylar knee unicondylar knee replacements, the third most
replacement is more likely to be performed on males. common indication was dislocation / subluxation
rather than pain. The incidence of revision for
TKRs with a monobloc polyethylene tibia consistently indications such as pain and aseptic loosening /
show some of the lowest crude revision rates, although lysis was lower for uncemented unicondylar than for
the numbers at risk in later years are small, so the cemented, but higher for dislocation / subluxation and
results must be interpreted with caution. Cemented periprosthetic fractures. Progression of osteoarthritis
TKRs that are unconstrained with a fixed bearing, elsewhere in the knee is also the fourth most common
as well as being the most common type of TKR, indication selected by surgeons for revision knee
consistently show low revision rates in comparison to replacement. The risk of revision for progressive
alternatives; crude revision rates are approximately one arthritis, aseptic loosening / lysis and pain were all
percentage point lower in comparison to cemented higher for UKRs than TKRs, but the risk of revision for
unconstrained TKRs with a mobile bearing and infection was lower.
cemented TKRs that are posterior stabilised, with either
a fixed or mobile bearing at ten years. Infection accounts for the majority of the two-stage
revision procedures performed. Approximately 8% of
Age and gender influence the risk of revision surgery, revisions for infection that have been recorded in the
with younger patients and males being more likely registry to date have been single-stage procedures,
to undergo revision; and it has previously been indicating low usage and take-up of this technique in
felt that this may explain the higher revision rates the treatment of knee prosthetic joint infection. The
observed in UKR. We present results divided by soft tissue envelope makes single-stage knee revision
gender and age group and these show the risk of surgery potentially more challenging than that in the
revision of a cemented unicondylar knee replacement hip, which may explain the differences in utilisation of a
is at least two times higher in males and 2.4 times single-stage approach.
higher in females at ten years than a cemented TKR.
The distinction of uncemented unicondylar knee The risk of re-revision following a revision procedure
replacements shows that revision rates are lower than is higher than for the risk of revision of a primary TKR
for cemented unicondylar replacements but remain across all types of knee replacement. The risk of re-
higher than for cemented TKR. The risk of revision revision of a revised patellofemoral replacement is
of a patellofemoral replacement is at least 2.9 times slightly lower than the other types of knee, with the
higher in males and females than a cemented TKR rest being broadly similar. This suggests that caution
across all age groups at ten years and the results of should be exercised when proposing that a UKR may
multicompartmental knee replacements show similarly be considered as an interim procedure or a lesser
intervention than a TKR, as the crude re-revision rates
are worse than the revision rates for primary TKR,
and are broadly similar regardless of the type of the
knee replacement implanted at the primary procedure.
We consider this to be an area that requires further
research to fully explore the risk of revision in light of
the different demographics in these groups. The risk
of re-revision is higher for those revised after a shorter
period of time following the primary and is associated
with the specific indication for revision. This suggests
that not all of the processes that lead to revision are
the same and that some are more aggressive than
others with consequences beyond the initial revision.
Knee replacement remains a safe procedure with

low rates of peri-operative mortality. The rates of
mortality are higher for males than those for females.
The average age of a patient undergoing a TKR is
approximately 70 years, just over 55% of males and
45% of females in the 70 to 74 age bracket will have
died within 15 years of their knee replacement. This
means that for the average patient undergoing a knee
replacement, their knee replacement should last
them for the rest of their life, without the need for
revision surgery.
3.4 Outcomes after
ankle replacement
3.4.1 Overview of primary ankle
replacement surgery
In this section of the report, we look at revision and
mortality for all primary ankle operations submitted to
the registry from 1 January 2010 up to 31 December
2020. There were, after data cleaning, 7,084
primary ankle operations available for analysis on
6,744 patients. A total of 340 patients had bilateral
operations (nine had both sides operated on the same
date), which can be seen in the patient flow diagram in
Figure 3.A1.
National Joint Registry | 18th Annual Report | Ankles
Figure 3.A1 Ankle cohort flow diagram.
Ankle procedures recorded by the NJR

N=8,440
Non−consenting procedures N=231
Non−traced procedures N=88
Invalid IDs N=0

N=8,121
*Procedures prior to 2010 N=14

N=8,107

N=8,107
Northern Ireland N=119
Isle of Man N=0

N=7,988
/ Gender / ASA grade / Indications
/ Unit / Prostheses used N=8
Duplicate same day & type revisions N=0
Unique procedures
N=7,980
Procedures (13 ankles) with
an inconsistent operative pattern N=27
Procedures (7,511 ankles)

N=7,953
*All revision procedures N=869

*Of which, ankles where first recorded
procedure is a revision N=481
Primary procedures
(Revision analyses)
N=7,084
Bilateral primary procedures (same day) N=9
N=7,075
* Reasons not necessarily mutually exclusive
The median age at primary surgery was 69 years listed in the component data. Cement was listed in
(IQR 62 to 75 years), with an overall range of 17 to 327 (4.6%) of primary procedures. Of all total ankle
97 years. More procedures were performed in men replacement (TAR) procedures, 184 (2.6%) were
(59.7%) than in women. defined as unconfirmed. Procedures were defined
as unconfirmed when they either had insufficient
All ankle replacement brands recorded in the registry elements to form a coherent construct or they
are uncemented implants, but cement can be used contained custom-made prostheses.
occasionally by surgeons in circumstances such as
poor bone stock or low demand patients. Of the Figure 3.A2 illustrates the temporal changes in fixation
7,084 primary procedures, a total of 6,757 (95.4%) of primary ankle replacements.
procedures were implanted without cement being
Figure 3.A2 Fixation by year of primary ankle replacement.
100
90
80
70
60
50
40
30
20
10
0
2010
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020
Year of primary
TAR (no cement listed) TAR (cement listed) Unconfirmed TAR
Figure 3.A3 and Figure 3.A4 show the yearly number of Figure 3.A3 shows the volume of primary ankle
primary ankle replacements performed for all indications replacements recorded in the registry increasing
and ankle replacements stratified by fixed and mobile since 2015 (except for 2020 due to the impact of
bearings, please note the difference in scale of the COVID-19). The majority of additional procedures
y-axis between each plot. Each bar in the figure is were contributed to the registry by higher volume
further stratified by the volume of procedures that the ankle surgeons i.e. surgeons who perform more than
surgeon conducted in that year, and when procedures 13 TAR procedures annually. Figure 3.A4 overleaf
are stratified by fixed and mobile bearings the volume illustrates that the expansion of TAR procedures has
of procedures is calculated separately. For example, if largely been of a fixed bearing design and that the use
a surgeon performed 25 primary ankle replacements of mobile bearing has steadily been decreasing. Many
procedures, their procedures would have contributed to of the changes in bearing use are due to the voluntary
the grey sub-division in Figure 3.A3. If those procedures withdrawal of the Mobility implant in 2014 and the
consisted of 12 fixed bearings and 13 mobile bearings, introduction of the Infinity in the same year.
those procedures would be represented by green and
purple bars respectively (Figure 3.A4).
Figure 3.A3 Frequency of primary ankle replacements, bars stacked by volume per consultant
per year.
800

600
Frequency (N=)
400
200
0 2010 2012 2014 2016 2018 2020

2011 2013 2015 2017 2019
1≤N≤2 3≤N≤4 5≤N≤6 7≤N≤12 13≤N≤24 25≤N≤48
Figure 3.A4 Frequency of primary ankle replacements stratified by fixed and mobile bearings, bars
stacked by volume per consultant per year.
Fixed Mobile
600
Frequency (N=)
400
200
0 10 12 14 16 18 20 10 12 14 16 18 20
11 13 15 17 19 11 13 15 17 19
1≤N≤2 3≤N≤4 5≤N≤6 7≤N≤12 13≤N≤24 25≤N≤48
Table 3.A1 Descriptive statistics of ankle procedures performed by consultant and unit by year of surgery.
Number of primary Year of surgery
replacements
during each year 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020
Number of procedures
in year 403 523 582 558 546 620 735 777 882 993 465
Units (N) 104 127 145 134 138 143 144 146 149 160 125
Mean number of primary
3.9 4.1 4.0 4.2 4.0 4.3 5.1 5.3 5.9 6.2 3.7
replacements per unit

Median (IQR) number
2 2 2 2 2 2 2 3 3 3 2
of any primary
(1 to 4) (1 to 5) (1 to 4) (1 to 5) (1 to 4) (1 to 5) (1 to 6.5) (1 to 6) (1 to 7) (2 to 8) (1 to 5)
replacements per unit
Units who entered ≥10
10 9 13 12 11 10 20 18 24 31 9
operations (N)
Units who entered ≥20
3 3 4 4 4 6 7 6 8 6 2
operations (N)
Consultants providing
107 126 143 133 126 142 137 141 148 155 116
operation (N)
Mean number of primary
replacements per 3.8 4.2 4.1 4.2 4.3 4.4 5.4 5.5 6.0 6.4 4.0
consultant
Median (IQR) number
of any primary 2 3 2 3 3 2 3 3 3.5 5 2
replacements per (1 to 4) (2 to 5) (1 to 5) (1 to 5) (2 to 5) (1 to 6) (2 to 8) (1 to 8) (2 to 8) (2 to 9) (1 to 5.5)
consultant
Consultants who entered
10 10 12 13 10 16 20 27 32 36 8
≥10 operations (N)
Consultants who entered
2 3 2 2 2 4 5 7 6 5 1
≥20 operations (N)
Table 3.A1 shows an increasing number of annually in 2010 and this number slowly increased to 3.2% of
reported cases over the ten year observation period. consultants by 2019, although in 2020 only 0.9% of
This could represent improved compliance or the all consultants performed more than 20 primary ankle
reporting of a true increase in caseload. replacements. The percentage of units submitting 20
or more ankle primary operations per year does not
A total of 282 consultants carried out the 7,084 exceed 5% (2018) (1.6 % in 2020). Of the 276 units
reported primary procedures over the ten year who submitted data to the registry, 11 (4%) carried out
period. The annual mean number of procedures 20 or more procedures since the start of ankle data
per consultant was 3.8 in 2010, and has gradually collection in 2010. The mean number of primary ankle
increased to 6.4 in 2019, although reduced to 4.0 in replacements per unit per year rose steadily from 3.9
2020 due to COVID-19. Only 1.9% of all consultants in 2010 to 6.2 in 2019, although this fell to 3.7 in 2020
performed more than 20 primary ankle replacements due to COVID-19.
220
Table 3.A2 Number and percentage of primary ankle replacements by brand.
Number (%) of each brand, for each year of operation

Number of
Brand primaries (%) 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020
Akile 45 (0.6) 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0) 4 (0.6) 9 (1.2) 12 (1.5) 11 (1.2) 8 (0.8) <4 (0.2)
Box 796 (11.2) 22 (5.5) 27 (5.2) 45 (7.7) 51 (9.1) 81 (14.8) 134 (21.6) 126 (17.1) 109 (14.0) 102 (11.6) 80 (8.1) 19 (4.1)
CCI <4 (0.0) 0 (0.0) 0 (0.0) <4 (0.2) 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0)
Cadence 59 (0.8) 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0) <4 (0.4) 6 (0.8) 15 (1.7) 24 (2.4) 11 (2.4)
Hintegra 296 (4.2) 9 (2.2) 17 (3.3) 35 (6.0) 67 (12.0) 47 (8.6) 54 (8.7) 33 (4.5) 9 (1.2) 14 (1.6) 11 (1.1) 0 (0.0)
Inbone 124 (1.8) 0 (0.0) 0 (0.0) <4 (0.3) 4 (0.7) 16 (2.9) <4 (0.5) 24 (3.3) 22 (2.8) 26 (2.9) 17 (1.7) 10 (2.2)
Inbone[Talar]
198 (2.8) 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0) 5 (0.9) 16 (2.6) 28 (3.8) 31 (4.0) 35 (4.0) 49 (4.9) 34 (7.3)
Infinity[Tibial]
Infinity 2,120 (29.9) 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0) 28 (5.1) 95 (15.3) 212 (28.8) 377 (48.5) 489 (55.4) 616 (62.0) 303 (65.2)
Infinity[Talar]
<4 (0.0) 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0) <4 (0.1) <4 (0.2) 0 (0.0)
Inbone[Tibial]
Mobility 1,117 (15.8) 252 (62.5) 295 (56.4) 284 (48.8) 201 (36.0) 85 (15.6) 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0)
Rebalance 61 (0.9) 0 (0.0) 4 (0.8) 13 (2.2) 13 (2.3) 6 (1.1) 4 (0.6) 13 (1.8) 7 (0.9) <4 (0.1) 0 (0.0) 0 (0.0)
Salto 323 (4.6) 22 (5.5) 29 (5.5) 40 (6.9) 45 (8.1) 55 (10.1) 55 (8.9) 44 (6.0) 9 (1.2) 11 (1.2) 9 (0.9) 4 (0.9)
Star 656 (9.3) 14 (3.5) 28 (5.4) 30 (5.2) 34 (6.1) 58 (10.6) 74 (11.9) 84 (11.4) 100 (12.9) 95 (10.8) 88 (8.9) 51 (11.0)
Trabecular
6 (0.1) 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0) 5 (0.7) 0 (0.0) <4 (0.1) 0 (0.0) 0 (0.0)
Metal Total
Vantage 19 (0.3) 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0) 16 (1.6) <4 (0.6)
Zenith 1,076 (15.2) 77 (19.1) 109 (20.8) 124 (21.3) 131 (23.5) 151 (27.7) 159 (25.6) 109 (14.8) 61 (7.9) 73 (8.3) 58 (5.8) 24 (5.2)
Unconfirmed 184 (2.6) 7 (1.7) 14 (2.7) 8 (1.4) 12 (2.2) 14 (2.6) 22 (3.5) 45 (6.1) 34 (4.4) 8 (0.9) 15 (1.5) 5 (1.1)
Total 7,084 (100) 403 (100) 523 (100) 582 (100) 558 (100) 546 (100) 620 (100) 735 (100) 777 (100) 882 (100) 993 (100) 465 (100)
Table 3.A2 shows the number of replacements by In 2020, the three most common brands were
implant brand and year of primary operation. The Infinity[Tal:Tib] (65.2%), Star[Tal:Tib] (11.0%) and
most frequently used brand is the fixed bearing, Inbone[Tal]Infinity[Tib] (7.3%). It was not possible to
Infinity[Tal:Tib] (Stryker), which represented 65.2% of identify the brand implanted in five procedures in 2020.
primary ankle replacements performed in 2020. The
use of this brand has risen steeply from its introduction 3.4.2 Revisions after primary
in 2014. ankle surgery
We are identifying when components, within primary A total of 339 out of the 7,084 primary procedures
ankle replacements, come from different brands had a linkable A2 MDS form completed to indicate
and/or manufacturers. There are no examples of a revision before the end of 2020. The first revisions
mix and match between manufacturers within ankle shown here include 43 conversions to arthrodesis,
replacements. The Infinity and Inbone implants, 238 single-stage procedures, 52 two-stage
both manufactured by Stryker, were designed to be procedures, six DAIRs, four with modular exchange
interchangeable with a matched articulating surface. and two without. No amputations have been recorded,
This combination represented 7.3% of primary ankle and, given the low rate reported for conversion to
replacements in 2020. Prior to the introduction of arthrodesis, we believe that these small numbers are
the Infinity, the Mobility (DePuy) had been the market likely to be a reflection of under-reporting.
leader before it was voluntarily withdrawn.
Table 3.A3 KM estimates of cumulative revision (95% CI) of primary ankle replacement, by gender and age.
Age at primary Number of Time since primary

(years) primaries 1 year 3 years 5 years 7 years 10 years

All cases 7,084 0.76 (0.58-1.00) 3.23 (2.80-3.71) 5.80 (5.16-6.52) 7.29 (6.50-8.16) 8.52 (7.55-9.60)
Female 2,855 0.84 (0.56-1.26) 3.59 (2.91-4.43) 6.31 (5.31-7.49) 8.00 (6.75-9.47) 9.41 (7.88-11.22)
<65 1,052 0.89 (0.46-1.70) 4.87 (3.64-6.50) 8.97 (7.10-11.30) 11.25 (8.98-14.05) 12.58 (9.96-15.81)
65 to 74 1,123 0.84 (0.44-1.61) 3.58 (2.55-5.01) 6.10 (4.61-8.06) 8.09 (6.14-10.62) 10.00 (7.49-13.29)
≥75 680 0.76 (0.32-1.82) 1.55 (0.80-2.99) 2.04 (1.12-3.71) 2.04 (1.12-3.71) 2.66 (1.39-5.04)
Male 4,229 0.71 (0.50-1.03) 2.97 (2.46-3.60) 5.45 (4.66-6.38) 6.77 (5.81-7.88) 7.86 (6.67-9.24)
<65 1,343 0.85 (0.47-1.54) 4.44 (3.38-5.83) 7.50 (5.95-9.43) 8.93 (7.13-11.16) 10.12 (7.97-12.79)
65 to 74 1,775 0.64 (0.35-1.15) 2.65 (1.95-3.61) 5.32 (4.16-6.80) 7.22 (5.72-9.09) 8.43 (6.60-10.74)
≥75 1,111 0.66 (0.32-1.38) 1.64 (0.98-2.72) 2.94 (1.90-4.52) 2.94 (1.90-4.52) 3.43 (2.15-5.45)
Note: Arthrodesis and amputation revision procedures may be under-reported in the registry.
Figure 3.A5 KM estimates of cumulative revision of primary ankle replacement (shaded area indicates
point-wise 95% CI).
10
8
0
0 1 2 3 4 5 6 7 8 9 10
Years since primary
Numbers at risk
7,084 6,519 5,406 4,438 3,580 2,816 2,200 1,689 1,180 690 279
Figure 3.A5 and Table 3.A3 show the overall estimated variations between brands were observed for later
cumulative percentage probability of (first) revision. post-operative periods, with rates varying from 2.76
Results are also stratified by gender and age. (95% CI 1.62-4.69) to 8.79 (95% CI 6.72-11.45) at
five years post-operation. The large relative differences
Table 3.A4 and Figure 3.A6 on page 224, show the between the lowest and highest rates seem to be
estimated cumulative percentage probability of (first) related to the implant brand and are unlikely to be
revision by implant brand with at least 250 uses. Rates entirely due to patient age and gender case mix. At
are not reported when there are less than ten primary ten years post-operation, the 95% confidence intervals
procedures at risk of revision for the considered time- are large, overlapping each other, and no robust
period. At one year post-operation rates of revision comparison between brands can be performed until
were heterogeneous between brands, varying from the size of the cohort becomes larger.
0.57 (95% CI 0.26-1.26) to 1.56 (0.65-3.70). Larger
Table 3.A4 KM estimates of cumulative revision (95% CI) of primary ankle replacement, by brand.
Number of Age Male Time since primary

Brand primaries Median (IQR) (%) 1 year 3 years 5 years 7 years 10 years
Box 796 67 (60 to 73) 65 1.28 (0.69-2.36) 4.90 (3.54-6.77) 8.79 (6.72-11.45) 11.43 (8.55-15.20) 11.43 (8.55-15.20)
Hintegra 296 70 (63 to 75) 66 0.68 (0.17-2.67) 2.83 (1.43-5.59) 4.87 (2.85-8.26) 6.44 (3.96-10.38)
Infinity 2,120 69 (62 to 75) 59 0.67 (0.39-1.15) 1.81 (1.25-2.63) 2.94 (1.93-4.45)
Mobility 1,117 68 (61 to 75) 56 0.81 (0.42-1.55) 4.55 (3.47-5.96) 8.42 (6.91-10.24) 10.13 (8.46-12.11) 11.10 (9.32-13.20)
Salto 323 69 (62 to 74) 59 1.56 (0.65-3.70) 3.53 (1.97-6.28) 5.32 (3.29-8.56) 5.77 (3.62-9.15) 7.66 (4.23-13.66)
Star 656 69 (63 to 76) 66 0.82 (0.34-1.96) 1.93 (1.07-3.47) 2.76 (1.62-4.69) 4.05 (2.28-7.15)
Zenith 1,076 69 (63 to 75) 58 0.57 (0.26-1.26) 4.21 (3.13-5.66) 6.44 (5.02-8.23) 7.49 (5.90-9.50) 8.31 (6.53-10.55)
Note: Rates are not reported when there are less than ten primary procedures at risk of revision for the considered time period.
Note: Brands with less than 250 procedures are not reported.
Note: Arthrodesis and amputation revision procedures may be under-reported in the registry.
223
Figure 3.A6 KM estimates of cumulative revision of primary ankle replacement by brand. Blue italics
in the numbers at risk table signify that fewer than 250 cases remained at risk at these time points.
15
12
9
0
0 1 2 3 4 5 6 7 8 9 10
Years since primary
Zenith 1,076 1,037 943 848 764 644 496 367 242 141 61
Star 656 594 499 402 300 218 147 94 61 32 7
Salto 323 313 295 279 261 219 168 116 76 44 18
Mobility 1,117 1,100 1,064 1,031 993 954 919 821 639 398 164
Infinity 2,120 1,787 1,166 678 309 108 25
Hintegra 296 292 278 259 243 205 146 102 45 16 5
Box 796 766 665 549 428 308 190 118 75 40 19
Table 3.A5 Indications for the first revisions following primary ankle replacement.
Note: These are not mutually exclusive.
Number of revisions per 100
Indication Total number revised prosthesis-years (95% CI)
Infection 92 0.28 (0.23-0.35)
Aseptic loosening 157 0.49 (0.42-0.57)
Aseptic loosening of tibial component
40 0.12 (0.09-0.17)
only
Aseptic loosening of talar component
48 0.15 (0.11-0.20)
only
Aseptic loosening of both tibial and
69 0.21 (0.17-0.27)
talar components
Lysis 65 0.20 (0.16-0.26)
Lysis of tibial component only 14 0.04 (0.03-0.07)
Lysis of talar component only 25 0.08 (0.05-0.11)

Lysis of both tibial and talar
26 0.08 (0.05-0.12)
components
Malalignment 60 0.19 (0.14-0.24)
Implant fracture 12 0.04 (0.02-0.07)
Implant fracture of tibial component
0 0
only
Implant fracture of talar component
<4 0.01 (0.00-0.02)
only
Implant fracture of meniscal
8 0.02 (0.01-0.05)
component only
Implant fracture of tibial and talar
<4 0.01 (0.00-0.02)
components
Meniscal insert dislocation 11 0.03 (0.02-0.06)
Wear of polyethylene component 34 0.11 (0.08-0.15)
Component migration/dissociation 25 0.08 (0.05-0.11)
Pain 70 0.22 (0.17-0.27)
Stiffness 36 0.11 (0.08-0.15)
Soft tissue impingement 30 0.09 (0.06-0.13)
Other indication for revision 43 0.13 (0.10-0.18)
Note: Two revision procedures recorded no reason for that revision and were removed from the analysis.
Note: In MDSv4 pain was referred to as Pain (undiagnosed) and from MDSv6 onwards pain is referred to as Unexplained Pain.
Table 3.A5 shows the indications for revision of ankle for an indication of lysis, 40.0% had lysis of both
replacements, with aseptic loosening and infection as tibial and talar components. Of the 12 revisions for
the most commonly cited indications. implant fracture, eight (66.7%) were performed for
a fractured meniscal insert and two (16.7%) were
Of the revisions for infection, 23 (25.0%) were performed to treat implant fracture of both tibial and
recorded as having a high suspicion of infection (e.g. talar components.
pus or confirmed micro) and the remaining revisions
for infection had a low suspicion (awaiting micro/ There is concern that there may be under-reporting of
histo). Out of the 157 revisions for aseptic loosening, revisions of ankle replacement, in particular when the
43.9% were performed because of loosening of both revision is to an ankle arthrodesis or amputation.
the tibial and talar components. Of patients revised
The NJR asks surgeons and those responsible for adding another implant to another part of the joint.
healthcare delivery to ensure that when primary and For the analyses of surgeon performance, hospital
revision joint replacement procedures of the hip, performance and implant performance, debridement
knee, ankle, elbow or shoulder are performed, that and implant retention without implant exchange is
the relevant MDS form is completed and data entered currently excluded.
into the registry. This is a requirement mandated by
the Department of Health and Social Care. For the 3.4.3 Mortality after primary
purposes of the annual report, revision procedures ankle replacement
include any addition, removal or modification of the
implants and procedures such as debridement and In this analysis, the second of each of the nine (same
implant retention with or without implant exchange, day) bilateral procedures were excluded. Among the
excision arthroplasty, amputation and conversion remaining 7,075, a total of 556 patients had died
to arthrodesis. For the avoidance of confusion, before the end of 2020, 189 of these were female
completing a revision MDS form is also mandatory and 367 were male.
for a procedure involving modification of a joint by
Figure 3.A7 KM estimates of cumulative mortality after primary ankle replacement (shaded area
indicates point-wise 95% CI).
25
20
15
10
0
0 1 2 3 4 5 6 7 8 9 10
Years since primary
Numbers at risk
7,075 6,559 5,526 4,602 3,777 3,024 2,382 1,835 1,287 762 313
Table 3.A6 KM estimates of cumulative mortality (95% CI) after primary ankle replacement, by gender and age.
Age at Time since primary
primary Number of
(years) primaries 30 days 90 days 1 year 3 years 5 years 7 years 10 years
All cases 7,075 0.07 (0.03-0.17) 0.16 (0.09-0.28) 0.79 (0.61-1.03) 3.01 (2.60-3.48) 6.41 (5.74-7.16) 10.95 (9.95-12.04) 21.71 (19.55-24.08)
Female 2,853 0.04 (0.00-0.25) 0.14 (0.05-0.38) 0.66 (0.41-1.04) 2.49 (1.93-3.21) 5.25 (4.32-6.36) 9.41 (8.00-11.07) 17.26 (14.47-20.52)
<65 1,050 0 0.19 (0.05-0.77) 0.39 (0.15-1.04) 1.19 (0.66-2.15) 2.37 (1.48-3.79) 4.97 (3.40-7.23) 7.47 (5.27-10.54)
65 to 74 1,123 0.09 (0.01-0.63) 0.18 (0.05-0.72) 0.65 (0.31-1.35) 2.13 (1.38-3.30) 4.56 (3.26-6.36) 8.07 (6.07-10.69) 14.35 (10.45-19.55)
≥75 680 0 0 1.08 (0.52-2.26) 5.24 (3.63-7.52) 11.28 (8.59-14.73) 19.28 (15.33-24.10) 39.20 (30.47-49.40)
Male 4,222 0.09 (0.04-0.25) 0.17 (0.08-0.35) 0.88 (0.64-1.22) 3.36 (2.81-4.01) 7.20 (6.29-8.24) 12.00 (10.66-13.50) 25.11 (22.02-28.55)
<65 1,342 0 0 0.08 (0.01-0.54) 1.43 (0.88-2.33) 3.29 (2.31-4.68) 4.44 (3.14-6.25) 9.98 (6.54-15.07)
65 to 74 1,771 0.17 (0.05-0.52) 0.23 (0.09-0.60) 0.93 (0.57-1.51) 2.78 (2.05-3.75) 6.21 (4.95-7.77) 9.61 (7.85-11.73) 18.47 (14.90-22.77)
≥75 1,109 0.09 (0.01-0.64) 0.28 (0.09-0.85) 1.79 (1.14-2.79) 6.77 (5.29-8.64) 13.92 (11.53-16.74) 26.02 (22.29-30.25) 59.42 (50.25-68.80)
Note: Some patients had operations on the left and right side on the same day. The second of bilateral operations performed on the same day were excluded.
227
Figure 3.A7 and Table 3.A6 on the previous pages 2010 to 2019, with an expected fall off in numbers in
show the estimated cumulative percentage probability 2020 due to COVID-19. Only 7.2% of units conducting
of death at different times after surgery, by gender and ankle replacements performed more than ten per
age at primary. Male patients and patients of older age year in 2020 and, in the same year, just 1.6% of units
were more likely to have died. performed more than 20 primary procedures. BOFAS
encourages surgeons to pool resources and create
3.4.4 Conclusions networks, where practicable, to ensure the sharing
of best practice in the achievement of the highest
Compared to the other joint types included in the standards of care and outcome quality for patients.
annual report, primary ankle replacement is a low
volume procedure, and linked revisions are even The cumulative percentage probability of 90-day
lower. It is likely that there is significant under-reporting mortality following primary ankle surgery is very low
of revision to arthrodesis procedures, or revision to (0.16% (95% CI 0.09-0.28)) and the cumulative
amputation, making outcome analysis difficult. percentage of revision at ten years following a primary
ankle replacement is found to be 8.52% (95% CI 7.55-
Since the withdrawal of the Mobility implant in 9.60). Substantial heterogeneity in the rates of revision
2014, the fixed bearing Infinity implant has rapidly was observed between the implant brands used in
gained popularity to become the market leader and primary ankle replacement surgery.
survivorship data is encouraging at present.
Although there has been a trend towards an increasing

volume of replacements by unit, the mean number per
unit has only risen from 3.9 to 6.2 per year between
3.5 Outcomes after
elbow replacement
3.5.1 Overview of primary elbow radial head replacement. We conducted an extended
review of the component labels reported on the
replacement surgery primary elbow (E1) MDS form. Our analysis has been
In this section we detail the primary elbow able to identify total replacements with a radial head
replacements entered into the registry since recording replacement (n=29) and investigate inconsistencies
began (1 April 2012) up to the end of 31 December between the type of procedure reported on the MDS
2020. Data on linked first revision episodes and form and the component label data uploaded to
linked mortality data are presented. Primary elbow the registry. Procedures where the reported type of
replacement in this section refers to total replacement surgery did not match the components listed on the
(with or without radial head replacement), distal MDS form are classified as unconfirmed in the elbow
humeral hemiarthroplasty, lateral resurfacing and section of the report.
National Joint Registry | 18th Annual Report | Elbows
Figure 3.E1 Elbow cohort flow diagram.
Elbow procedures recorded by the NJR

N=6,980
Non−consenting procedures N=332
Invalid IDs N=0

N=6,592

N=6,587

N=6,587
Isle of Man N=0

N=6,486
/ Gender / ASA grade / Indications
/ Unit / Prostheses used N=9
Unique procedures
N=6,476
Procedures (8 elbows) with
Procedures (5,938 elbows)

N=6,460

*Of which, elbow procedures where first recorded
procedure is a revision N=1,129
Primary procedures
(Revision analyses)
N=5,043
Bilateral procedures (same day) N=10
N=5,033
A total of 5,043 primary replacements were available replacement and distal humeral hemiarthroplasties, or
for analysis for a total of 4,861 patients (Figure 3.E1). a combination of these factors. There is a decrease
Of these patients, 182 had documented elbow from 2019 to 2020 due to the impact of COVID-19.
replacements on both left and right sides, and in ten
patients these were both performed on the same Table 3.E1 provides a breakdown by the stated
day (bilateral). type of replacement. Of all procedures, including
the unconfirmed, 68.3% were classified as a total
The majority of replacements were performed on replacement. A total of 454 (9%) primary elbow
women (69.1%) and the median age at the time of replacements had an unconfirmed status.
primary operation was 67 years (IQR 56 to 76), with
an overall range of 14 to 99 years. Cement was listed Table 3.E2 (page 234) details the type of primary
in the component data in 66.0% of the primary operation in each year and we show that 2,135
elbow procedures. (42.3%) elbow replacements were carried out for
acute trauma. These have been separated from
Table 3.E1 shows that the annual number of primary the remaining 2,908 cases performed for elective
elbow replacements entered into the registry has indications in the rest of this section. Nearly half
increased since 2012. While the increase in the early (49.1%) of the elbow procedures performed for trauma
years is in part due to improvement in data capture, were confirmed radial head replacements.
the consistent increase observed year after year from
2015 to 2019 mostly reflects an increase in the volume
of procedures, improved reporting of radial head
Table 3.E1 Number of primary elbow replacements by year and percentage of each type of procedure.
Number Year of primary

of 2012 2013 2014 2015 2016 2017 2018 2019 2020
primaries N (%) N (%) N (%) N (%) N (%) N (%) N (%) N (%) N (%)
All cases 5,043 258 (100.0) 451 (100.0) 449 (100.0) 545 (100.0) 577 (100.0) 658 (100.0) 710 (100.0) 834 (100.0) 561 (100.0)
Confirmed elbow
4,589 198 (76.7) 370 (82.0) 408 (90.9) 486 (89.2) 515 (89.3) 600 (91.2) 661 (93.1) 806 (96.6) 545 (97.1)
replacements
Total elbow
3,031 170 (65.9) 329 (72.9) 351 (78.2) 390 (71.6) 384 (66.6) 433 (65.8) 381 (53.7) 381 (45.7) 212 (37.8)
replacement
Total elbow
replacement
29 0 (0.0) 0 (0.0) <4 (0.7) <4 (0.2) <4 (0.5) <4 (0.3) 8 (1.1) 9 (1.1) <4 (0.5)
inc. radial head
replacement
Radial head
1,286 23 (8.9) 36 (8.0) 53 (11.8) 94 (17.2) 128 (22.2) 163 (24.8) 220 (31.0) 316 (37.9) 253 (45.1)
replacement
Lateral resurfacing 13 5 (1.9) 5 (1.1) <4 (0.2) 0 (0.0) 0 (0.0) <4 (0.2) 0 (0.0) 0 (0.0) <4 (0.2)
Distal humeral
230 0 (0.0) 0 (0.0) 0 (0.0) <4 (0.2) 0 (0.0) <4 (0.2) 52 (7.3) 100 (12.0) 76 (13.5)
hemiarthroplasty
Unconfirmed elbow
454 60 (23.3) 81 (18.0) 41 (9.1) 59 (10.8) 62 (10.7) 58 (8.8) 49 (6.9) 28 (3.4) 16 (2.9)
replacements
Unconfirmed total
383 53 (20.5) 78 (17.3) 36 (8.0) 54 (9.9) 53 (9.2) 49 (7.4) 40 (5.6) 14 (1.7) 6 (1.1)
elbow replacement
Unconfirmed radial
47 <4 (0.8) <4 (0.4) 5 (1.1) 5 (0.9) 7 (1.2) 6 (0.9) 5 (0.7) 9 (1.1) 6 (1.1)
head replacement
Unconfirmed lateral
12 5 (1.9) <4 (0.2) 0 (0.0) 0 (0.0) <4 (0.2) <4 (0.3) <4 (0.3) 0 (0.0) <4 (0.2)
resurfacing
Unconfirmed
distal humeral 12 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0) <4 (0.2) <4 (0.2) <4 (0.3) 5 (0.6) <4 (0.5)
hemiarthroplasty
Note: Elbow replacements with a mismatch between the type of procedure reported by the surgeon on the MDS form and the recorded component labels on the MDS form, or with no component data in the
record, are described as unconfirmed and classified according to the procedure type indicated by the surgeon on the MDS form.
233
234
Table 3.E2 Types of primary elbow procedures used in acute trauma and elective cases by year and type of primary operation.
Number Year of primary

of 2012 2013 2014 2015 2016 2017 2018 2019 2020
primaries N (%) N (%) N (%) N (%) N (%) N (%) N (%) N (%) N (%)
All cases 2,135 66 (100.0) 120 (100.0) 122 (100.0) 202 (100.0) 214 (100.0) 246 (100.0) 331 (100.0) 455 (100.0) 379 (100.0)
Confirmed elbow replacements 1,925 48 (72.7) 95 (79.2) 108 (88.5) 180 (89.1) 183 (85.5) 205 (83.3) 301 (90.9) 438 (96.3) 367 (96.8)
Total elbow replacement 673 33 (50.0) 65 (54.2) 62 (50.8) 104 (51.5) 85 (39.7) 81 (32.9) 77 (23.3) 85 (18.7) 81 (21.4)
Total elbow replacement inc. radial
<4 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0) <4 (0.2) 0 (0.0)
head replacement
Radial head replacement 1,048 15 (22.7) 30 (25.0) 46 (37.7) 75 (37.1) 98 (45.8) 123 (50.0) 182 (55.0) 261 (57.4) 218 (57.5)
Lateral resurfacing 0 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0)
Distal humeral hemiarthroplasty 203 0 (0.0) 0 (0.0) 0 (0.0) <4 (0.5) 0 (0.0) <4 (0.4) 42 (12.7) 91 (20.0) 68 (17.9)
Unconfirmed elbow replacements 210 18 (27.3) 25 (20.8) 14 (11.5) 22 (10.9) 31 (14.5) 41 (16.7) 30 (9.1) 17 (3.7) 12 (3.2)
Acute trauma
Unconfirmed total elbow replacement 163 16 (24.2) 24 (20.0) 9 (7.4) 19 (9.4) 26 (12.1) 34 (13.8) 25 (7.6) 6 (1.3) 4 (1.1)
Unconfirmed radial head replacement 37 <4 (3.0) <4 (0.8) 5 (4.1) <4 (1.5) 5 (2.3) 5 (2.0) <4 (0.9) 7 (1.5) 6 (1.6)
Unconfirmed lateral resurfacing <4 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0) <4 (0.4) <4 (0.3) 0 (0.0) 0 (0.0)
Unconfirmed distal humeral
8 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0) <4 (0.4) <4 (0.3) 4 (0.9) <4 (0.5)
hemiarthroplasty
All cases 2,908 192 (100.0) 331 (100.0) 327 (100.0) 343 (100.0) 363 (100.0) 412 (100.0) 379 (100.0) 379 (100.0) 182 (100.0)
Confirmed elbow replacements 2,664 150 (78.1) 275 (83.1) 300 (91.7) 306 (89.2) 332 (91.5) 395 (95.9) 360 (95.0) 368 (97.1) 178 (97.8)
Total elbow replacement 2,358 137 (71.4) 264 (79.8) 289 (88.4) 286 (83.4) 299 (82.4) 352 (85.4) 304 (80.2) 296 (78.1) 131 (72.0)
Total elbow replacement inc. radial

28 0 (0.0) 0 (0.0) <4 (0.9) <4 (0.3) <4 (0.8) <4 (0.5) 8 (2.1) 8 (2.1) <4 (1.6)
head replacement
Radial head replacement 238 8 (4.2) 6 (1.8) 7 (2.1) 19 (5.5) 30 (8.3) 40 (9.7) 38 (10.0) 55 (14.5) 35 (19.2)
Lateral resurfacing 13 5 (2.6) 5 (1.5) <4 (0.3) 0 (0.0) 0 (0.0) <4 (0.2) 0 (0.0) 0 (0.0) <4 (0.5)
Distal humeral hemiarthroplasty 27 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0) 10 (2.6) 9 (2.4) 8 (4.4)
Elective
Unconfirmed elbow replacements 244 42 (21.9) 56 (16.9) 27 (8.3) 37 (10.8) 31 (8.5) 17 (4.1) 19 (5.0) 11 (2.9) 4 (2.2)
Unconfirmed total elbow replacement 220 37 (19.3) 54 (16.3) 27 (8.3) 35 (10.2) 27 (7.4) 15 (3.6) 15 (4.0) 8 (2.1) <4 (1.1)
Unconfirmed radial head replacement 10 0 (0.0) <4 (0.3) 0 (0.0) <4 (0.6) <4 (0.6) <4 (0.2) <4 (0.5) <4 (0.5) 0 (0.0)
Unconfirmed lateral resurfacing 10 5 (2.6) <4 (0.3) 0 (0.0) 0 (0.0) <4 (0.3) <4 (0.2) <4 (0.3) 0 (0.0) <4 (0.5)
Unconfirmed distal humeral
4 0 (0.0) 0 (0.0) 0 (0.0) 0 (0.0) <4 (0.3) 0 (0.0) <4 (0.3) <4 (0.3) <4 (0.5)
hemiarthroplasty
Note: Elbow replacements with a mismatch between the type of procedure reported by the surgeon on the MDS form and the recorded component labels on the MDS form or with no component data in the record are
described as unconfirmed and classified according to the procedure type indicated by the surgeon on the MDS form.
Figure 3.E2 and Figure 3.E3 show the yearly number over the last five years (except for 2020 due to the
of primary elbow replacements performed for elective impact of COVID-19), with the number of surgeons
and acute trauma indications respectively. Elective performing one or two procedures annually falling.
and acute trauma procedures have been stratified by Elective radial head replacements are increasingly
total elbow replacements (with or without a radial head being recorded in the registry, however the majority
replacement), radial head replacements and distal of consultants only perform one or two procedures
humeral hemi-arthroplasty, please note the difference annually. Figure 3.E3 overleaf shows the volume of
in scale of the y-axis between each sub-plot. Each primary total elbow replacements staying relatively
bar in the figure is further stratified by the volume of constant over the last five years. In the last three years
procedures that the surgeon conducted in that year there has been an increasing proportion of primary
across both elective and acute trauma settings i.e. if total elbow replacements performed by higher volume
a surgeon performed 12 elective primary total elbow elbow surgeons i.e. those performing more than 13
replacement procedures and 12 acute trauma primary procedures a year. Radial head replacements for
total elbow replacement procedures, their annual total acute trauma have been steadily increasing in volume
volume would be 24 procedures. Those 24 procedures and the proportion of consultants performing three or
would contribute to the dark purple sub-division in more procedures per year has also been increasing,
both elective and acute trauma figures shown here. indicating a degree of specialisation among a minority
of consultants.
Figure 3.E2 shows that the volume of primary total
elbow replacements has marginally increased
Figure 3.E2 Frequency of primary elbow replacements within elective cases stratified by procedure type,
bars stacked by volume per consultant per year.
Total elbow replacement

(with or without radial head) Radial head replacement
200 30
150
20
100
50 10
Frequency (N=)

0 2012 2014 2016 2018 2020
0 2012 2014 2016 2018 2020
2013 2015 2017 2019 2013 2015 2017 2019
Distal humeral hemiarthroplasty
10
8
6
4
2
0 2012 2014 2016 2018 2020
2013 2015 2017 2019
1≤N≤2 3≤N≤4 5≤N≤6 7≤N≤12 13≤N≤24
Figure 3.E3 Frequency of primary elbow replacements within acute trauma cases stratified by procedure
type, bars stacked by volume per consultant per year.

(with or without radial head) Radial head replacement
60 150
40 100
20 50
Frequency (N=)
0 2012 2014 2016 2018 2020

0 2012 2014 2016 2018 2020
2013 2015 2017 2019 2013 2015 2017 2019
60
40
20
0 2012 2014 2016 2018 2020
2013 2015 2017 2019
1≤N≤2 3≤N≤4 5≤N≤6 7≤N≤12 13≤N≤24
Table 3.E3 Indications for main confirmed types of primary elbow replacements, by year and type of primary operation.
Acute Elective
trauma Number (%)* for each indication (amongst elective cases only)
Number Number Number
Year of of of cases of cases Inflammatory Trauma Essex Avascular Other
primary primaries (%) (%) Osteoarthritis arthropathy sequelae Lopresti necrosis indication
All cases 3,031 673 (22.2) 2,358 (77.8) 798 (33.8) 1,164 (49.4) 400 (17.0) 4 (0.2) 4 (0.2) 109 (4.6)
2012 170 33 (19.4) 137 (80.6) 44 (32.1) 65 (47.4) 28 (20.4) <4 (0.7) 0 (0.0) 7 (5.1)
2013 329 65 (19.8) 264 (80.2) 94 (35.6) 134 (50.8) 30 (11.4) <4 (0.4) <4 (0.4) 15 (5.7)
2014 351 62 (17.7) 289 (82.3) 105 (36.3) 146 (50.5) 38 (13.1) 0 (0.0) 0 (0.0) 15 (5.2)
2015 390 104 (26.7) 286 (73.3) 99 (34.6) 148 (51.7) 39 (13.6) 0 (0.0) <4 (0.7) 16 (5.6)
2016 384 85 (22.1) 299 (77.9) 100 (33.4) 150 (50.2) 52 (17.4) 0 (0.0) 0 (0.0) 12 (4.0)
2017 433 81 (18.7) 352 (81.3) 115 (32.7) 179 (50.9) 60 (17.0) <4 (0.3) <4 (0.3) 13 (3.7)
2018 381 77 (20.2) 304 (79.8) 103 (33.9) 159 (52.3) 50 (16.4) <4 (0.3) 0 (0.0) 10 (3.3)
2019 381 85 (22.3) 296 (77.7) 98 (33.1) 136 (45.9) 62 (20.9) 0 (0.0) 0 (0.0) 14 (4.7)

2020 212 81 (38.2) 131 (61.8) 40 (30.5) 47 (35.9) 41 (31.3) 0 (0.0) 0 (0.0) 7 (5.3)
All cases 1,286 1,048 238 43 <4 163 21 4 13
2012 23 15 8 <4 0 4 0 0 <4
Radial head replacement
2013 36 30 6 <4 0 4 0 0 0
2014 53 46 7 0 <4 5 <4 0 0
2015 94 75 19 4 0 14 0 <4 0
2016 128 98 30 5 0 24 <4 <4 <4
2017 163 123 40 6 0 27 4 0 4
2018 220 182 38 9 0 24 4 0 <4
2019 316 261 55 7 <4 41 4 <4 <4
2020 253 218 35 7 0 20 7 0 <4
All cases 230 203 27 5 <4 19 0 0 <4
2012 0 0 0 0 0 0 0 0 0
2013 0 0 0 0 0 0 0 0 0
2014 0 0 0 0 0 0 0 0 0
2015 <4 <4 0 0 0 0 0 0 0
2016 0 0 0 0 0 0 0 0 0
2017 <4 <4 0 0 0 0 0 0 0
2018 52 42 10 <4 <4 6 0 0 0
2019 100 91 9 <4 <4 8 0 0 0
2020 76 68 8 <4 0 5 0 0 <4
*Percentages are not presented where numbers are too few for meaningful percentages; please note the listed reasons are not mutually exclusive as more than one
reason could have been stated.
Note: Procedures with unconfirmed prostheses, confirmed lateral resurfacing and confirmed total elbow replacements including a radial head replacement were not
reported in this table.
Note: Distal humeral hemiarthroplasty started to be reported in MDSv7 released in June 2018.
Table 3.E3 describes the indications for the primary than one indication could have been provided. Only
operation separately by type of primary elbow one indication for surgery, as defined in Table 3.E3,
replacement. Primary operations with an unconfirmed was given for all 1,925 acute trauma cases with a
procedure type are excluded from this table. confirmed type of primary procedure. In 132 (5%)
of the 2,664 elective cases with a confirmed type of
Please note that the indications for primary elbow primary, more than one indication was given.
replacement are not mutually exclusive since more
238
Table 3.E4 Number of units and consultant surgeons (cons) providing primary elbow replacements during each year from the last three years, by region.
(a) All primary elbow replacements (including the confirmed and unconfirmed total, radial head, lateral resurfacing and distal humeral hemiarthroplasty replacements).
Year of primary
2018 2019 2020
Median Median Median Median Median Median
number of number of number of number of number of number of
Number Number primaries Number primaries Number Number primaries Number primaries Number Number primaries Number primaries
of of per unit of per cons. of of per unit of per cons. of of per unit of per cons.
Region primaries units (IQR) cons. (IQR) primaries units (IQR) cons. (IQR) primaries units (IQR) cons. (IQR)
All regions 710 172 2 (1 to 5) 231 5 (2 to 8) 834 172 3 (1 to 7) 249 6 (3 to 10) 561 138 2 (1 to 6) 213 4 (2 to 8)
East Midlands 77 15 5 (1 to 7) 30 7 (5 to 7) 65 12 3 (2 to 8.5) 24 6 (3 to 11.5) 42 8 3 (2.5 to 7.5) 18 4 (3 to 15)
East of England 72 13 4 (2 to 8) 17 8 (5 to 9) 70 13 2 (1 to 8) 18 10 (7 to 12) 41 9 6 (2 to 6) 17 6 (4 to 6)
London 62 24 2 (1 to 2.5) 30 2 (1 to 5) 88 25 3 (1 to 4) 33 3 (2 to 8) 82 27 2 (1 to 3) 36 2 (1 to 7)
North East 79 12 6 (4 to 9) 16 8.5 (5 to 10) 73 12 6.5 (4 to 8) 18 7.5 (7 to 9) 64 14 4 (1 to 7) 21 6 (4 to 8)
North West 121 27 2 (1 to 5) 42 4 (2 to 8) 145 27 2 (1 to 8) 40 8 (3 to 10) 64 16 2.5 (1.5 to 5.5) 32 3 (2 to 7)
South Central 29 5 6 (5 to 8) 11 6 (5 to 8) 38 6 7.5 (4 to 8) 13 8 (7 to 9) 33 7 3 (1 to 8) 10 4.5 (2 to 12)
South East
55 25 2 (1 to 3) 18 2.5 (1 to 4) 76 19 3 (1 to 7) 23 5 (3 to 8) 73 19 2 (1 to 6) 22 4 (2 to 11)
Coast
South West 56 15 2 (1 to 5) 22 5 (2 to 9) 99 15 3 (1 to 13) 24 13 (3 to 16) 59 8 6.5 (2.5 to 10) 20 7 (6 to 9)
Wales 37 9 3 (1 to 4) 6 5 (3 to 6) 43 12 3.5 (1.5 to 4) 10 4 (3 to 4) 9 5 1 (1 to 3) 4 2 (1 to 3)
West Midlands 49 13 3 (2 to 3) 19 3 (2 to 15) 54 14 3 (2 to 4) 26 3 (3 to 10) 34 11 3 (2 to 4) 16 3 (3 to 4)
Yorkshire and
73 14 4 (1 to 7) 20 5 (4 to 7.5) 83 17 3 (1 to 5) 20 4 (2 to 10) 60 14 2 (1 to 6) 17 5 (2 to 6)
the Humber
Note: Wales combines North, Mid and West, and South East regions.
Table 3.E4 Number of units and consultant surgeons (cons) providing primary elbow replacements during each year from the last three years, by region.
(b) All confirmed primary total elbow replacements (with or without radial head replacement).
Year of primary
2018 2019 2020
Median Median Median Median Median Median
number of number of number of number of number of number of
Number Number primaries Number primaries Number Number primaries Number primaries Number Number primaries Number primaries
of of per unit of per cons. of of per unit of per cons. of of per unit of per cons.
Region primaries units (IQR) cons. (IQR) primaries units (IQR) cons. (IQR) primaries units (IQR) cons. (IQR)
All regions 389 125 2 (1 to 4) 151 3 (2 to 6) 390 121 2 (1 to 4) 142 3 (2 to 6) 215 88 1 (1 to 3) 101 2 (1 to 5)
East Midlands 45 10 3 (2 to 5) 18 5 (3 to 5) 39 8 4 (2 to 6.5) 14 6 (3 to 7) 15 5 3 (1 to 3) 8 3 (2 to 5)
East of England 45 10 5 (4 to 6) 14 6 (5 to 6) 32 10 2.5 (1 to 5) 11 3 (2 to 6) 18 8 1.5 (1 to 3.5) 9 2 (1 to 4)
London 32 12 2 (1 to 3.5) 18 3.5 (2 to 8) 34 13 2 (1 to 3) 15 3 (2 to 5) 26 9 1 (1 to 2) 15 2 (1 to 9)
North East 28 11 2 (2 to 3) 12 2 (2 to 4) 32 11 3 (1 to 5) 11 4 (1 to 5) 26 11 1 (1 to 4) 12 3 (1 to 6)
North West 52 17 1 (1 to 3) 20 1 (1 to 3.5) 53 17 2 (1 to 3) 22 2 (1 to 4) 23 10 1.5 (1 to 3) 14 2 (1 to 3)
South Central 16 4 3.5 (2.5 to 5.5) 6 3.5 (3 to 4) 19 5 2 (2 to 6) 9 6 (2 to 6) 12 5 2 (1 to 3) 5 3 (3 to 5)
South East
31 17 2 (1 to 2) 12 2.5 (2 to 3.5) 30 11 2 (1 to 4) 11 4 (2 to 4) 28 11 1 (1 to 5) 8 2 (1 to 4)
Coast
South West 32 14 1.5 (1 to 4) 16 2.5 (1 to 4) 46 12 2.5 (1.5 to 5) 10 4 (2 to 7) 9 5 2 (1 to 2) 7 2 (1 to 3)
Wales 22 8 2.5 (1 to 4) 6 4 (3 to 6) 20 8 2 (1.5 to 3.5) 7 3 (2 to 4) 7 5 1 (1 to 1) 4 1 (1 to 2)
West Midlands 37 10 2 (1 to 3) 15 3 (2 to 15) 35 12 2 (1 to 3.5) 17 3 (2 to 5) 20 8 2 (1 to 2.5) 9 2 (2 to 8)
Yorkshire and
49 12 3 (1 to 5) 14 4 (2 to 11) 50 14 2 (1 to 4) 15 3 (2 to 6) 31 11 1 (1 to 4) 10 1.5 (1 to 4)
the Humber
Note: Wales combines North, Mid and West, and South East regions.
239
Over the last three years (from 2018 to 2020) 2,105 The median number of elbow replacements per unit
primary elbow replacements were entered into the and consultant has changed very little over the last
registry, of which 994 had confirmed components three years and remains around two to three per annum
consistent with a total elbow replacement (with or with up to 6.5 replacements per unit in the South West
without radial head replacement). region and as low as one replacement per unit in Wales
in 2020. These figures are subject to change, as some
On the previous pages Tables 3.E4 (a) and (b) show units may not have submitted all data for all 2020
the number of all types of elbow replacement by year procedures by the time of data analysis.
and NJR geographical region over this time period,
together with the number of units and consultants. Table 3.E5 below lists the brands used in elbow
A list of units within each NJR region is provided in replacement by confirmed procedure type, with sub-
the downloads section of reports.njrcentre.org.uk division by acute trauma and elective cases.
and further information can be found on
https://surgeonprofile.njrcentre.org.uk.
Table 3.E5 Brands used in elbow replacement by confirmed procedure type.
Number of
primaries Elective Acute trauma
All cases 3,031 2,358 673
Linked:
Coonrad Morrey 1,563 1,182 381
Discovery 777 615 162
GSB III 43 40 <4
Latitude EV Stem[Hum:Ulna] 179 151 28
Latitude EV Stem[Hum]Latitude EV Short
51 42 9
Stem[Ulna]
Latitude EV Stem[Hum]Latitude Legacy
<4 <4 0
Stem[Ulna]
Latitude Legacy Stem[Hum:Ulna] 31 25 6
Total elbow Latitude Legacy Stem[Hum]Latitude Legacy
38 31 7
replacement Short Stem[Ulna]
MUTARS Stem Cementless[Hum]MUTARS[Ulna] <4 <4 0
Nexel 240 170 70
Unlinked:
IBP 8 8 0
Latitude EV Stem[Hum:Ulna] 38 33 5
25 24 <4
Stem[Ulna]
Latitude Legacy Stem[Hum:Ulna] 9 9 0
Latitude Legacy Stem[Hum]Latitude Legacy
20 20 0
Short Stem[Ulna]
NES <4 <4 0
Note: Procedures of unconfirmed type are not reported in this table.

Note: [Hum]=Humeral, [Ulna]=Ulna, [Rad]=Radial Head, [LHR]=Lateral humeral resurfacing, [LRR]=Lateral radial resurfacing, [DHH]=Distal humeral hemiarthroplasty.
Table 3.E5 (continued)
Number of
primaries Elective Acute trauma
All cases 29 28 <4
Linked:
<4 <4 0
Stem[Ulna]Latitude (Legacy | EV)[Rad]
Latitude EV Stem[Hum]Latitude EV Stem[Ulna]
6 5 <4
Latitude (Legacy | EV)[Rad]
Latitude Legacy Stem[Hum]Latitude EV
<4 <4 0
Total elbow <4 <4 0
Short Stem[Ulna]Latitude (Legacy | EV)[Rad]
replacement
inc. radial head Unlinked:
replacement Latitude EV Stem[Hum]Latitude EV Short
<4 <4 0
6 6 0
Latitude (Legacy | EV)[Rad]
<4 <4 0
Latitude EV[Rad]
6 6 0
Short Stem[Ulna]Latitude (Legacy | EV)[Rad]

4 4 0
All cases 1,286 238 1,048
Bipolar:
Latitude (Legacy | EV)[Rad] <4 0 <4
RHS[Rad] 35 19 16
rHead Recon[Rad] 6 <4 <4
Monopolar:
Radial head Anatomic[Rad] 723 121 602
replacement Ascension[Rad] 79 21 58
Corin[Rad] 26 5 21
Evolve Proline[Rad] 276 43 233
ExploR[Rad] 104 16 88
Liverpool[Rad] 4 <4 <4
MoPyC[Rad] 9 <4 7
Uni-Radial Elbow[Rad] 6 <4 4
All cases 13 13 0
Lateral resurfacing LRE[LHR:LRR] 12 12 0
Uni-Elbow[LHR:LRR] <4 <4 0
All cases 230 27 203
Distal humeral
Latitude EV Stem[DHH] 213 23 190
hemiarthroplasty
Latitude Legacy Stem[DHH] 17 4 13
Note: Procedures of unconfirmed type are not reported in this table.

Note: [Hum]=Humeral, [Ulna]=Ulna, [Rad]=Radial Head, [LHR]=Lateral humeral resurfacing, [LRR]=Lateral radial resurfacing, [DHH]=Distal humeral hemiarthroplasty.
The top five constructs (Coonrad Morrey[Hum:Ulna], 3.5.2 Revisions after primary elbow
Discovery[Hum:Ulna], Nexel[Hum:Ulna], Latitude EV
Stem[Hum:Ulna], Latitude EV Stem[Hum]Latitude
replacement surgery
EV Short Stem[Ulna]) account for nearly 93.9% of We found that a total of 219 elbow primaries in the
total elbow replacements performed. All total elbow registry (48 acute trauma cases and 171 elective)
replacements with radial head replacement were had linked revision procedures recorded up to the
performed using the Latitude family of implants. end of 2020, including six excision procedures, 130
One implant, (RHS[Rad]), accounts for 83.3% of the single-stage revisions, nine DAIRs (seven with modular
bipolar radial head replacements and two implants, exchange and two without modular exchange) and 63
(Anatomic[Rad] and Evolve Proline[Rad]), account for stage one of a two-stage procedure.
81.4% of the monopolar radial head replacements.
Nearly all (92.3%) lateral resurfacing procedures have
been performed using the LRE[LHR:LRR] brand. The
Latitude EV Stem[DHH] was used for 92.6% of distal
humeral hemiarthroplasty procedures.
Table 3.E6 KM estimates of cumulative revision (95% CI) by primary elbow procedures for acute trauma and elective cases.
Age Time since primary

Number of (Median, Male
primaries IQR) (%) 1 year 2 years 3 years 4 years 5 years 6 years 7 years 8 years
All acute trauma and 1.32 2.74 4.13 5.39 5.97 7.04 7.53 7.76
5,043 67 (56 to 76) 31
elective cases (1.03-1.69) (2.29-3.28) (3.54-4.82) (4.67-6.22) (5.18-6.88) (6.09-8.14) (6.49-8.74) (6.64-9.06)
All acute trauma 1.09 2.03 2.38 2.90 3.29 3.89 3.89 3.89
2,135 66 (52 to 77) 30
cases (0.72-1.66) (1.46-2.81) (1.74-3.25) (2.12-3.96) (2.39-4.53) (2.76-5.47) (2.76-5.47) (2.76-5.47)
Total elbow 1.27 2.42 3.35 4.29 5.13 6.40 6.40 6.40
673 77 (71 to 83) 17
replacement (0.63-2.52) (1.44-4.07) (2.11-5.30) (2.77-6.60) (3.34-7.83) (4.14-9.85) (4.14-9.85) (4.14-9.85)
Total elbow
replacement
<4 79 (79 to 79) 0
inc. radial head
replacement
Radial head 0.30 0.59 0.59 0.89 0.89 0.89 0.89 0.89
1,048 53 (40.5 to 63) 42
replacement (0.10-0.95) (0.24-1.44) (0.24-1.44) (0.37-2.13) (0.37-2.13) (0.37-2.13) (0.37-2.13) (0.37-2.13)
Distal humeral 3.96 7.67
203 71 (65 to 79) 18
hemiarthroplasty (1.89-8.19) (3.84-15.01)
Acute trauma
Unconfirmed total 1.23 1.95 1.95 1.95 1.95 1.95 1.95
163 75 (66 to 83) 20
elbow replacement (0.31-4.85) (0.63-5.93) (0.63-5.93) (0.63-5.93) (0.63-5.93) (0.63-5.93) (0.63-5.93)
Unconfirmed radial 2.70 5.95 5.95 5.95 5.95
37 48 (37 to 59) 51
head replacement (0.39-17.68) (1.51-21.89) (1.51-21.89) (1.51-21.89) (1.51-21.89)
Unconfirmed lateral
<4 74.5 (74 to 75) 50
resurfacing
Unconfirmed
distal humeral 8 74 (63 to 80.5) 38
hemiarthroplasty
Note: Elbow replacements with a mismatch between the type of procedure reported by the surgeon on the MDS form and the recorded component labels on the MDS form or with no component data in the record
are described as unconfirmed and classified according to the procedure type indicated by the surgeon on the MDS form.
243
244
Age Time since primary

Number of (Median, Male
primaries IQR) (%) 1 year 2 years 3 years 4 years 5 years 6 years 7 years 8 years
1.46 3.17 5.07 6.68 7.33 8.59 9.24 9.54
All elective cases 2,908 68 (58 to 75) 31
(1.08-1.98) (2.56-3.91) (4.25-6.05) (5.68-7.83) (6.26-8.58) (7.32-10.05) (7.86-10.85) (8.06-11.27)
Total elbow 1.14 2.97 4.74 6.34 7.16 8.45 8.86 9.28
2,358 69 (60 to 76) 29
replacement (0.78-1.67) (2.33-3.79) (3.87-5.80) (5.28-7.61) (5.98-8.56) (7.06-10.11) (7.37-10.62) (7.63-11.25)
Total elbow
replacement 3.57 8.93
28 64 (55 to 71) 36
inc. radial head (0.51-22.76) (2.24-32.08)
replacement
Radial head 3.26 3.26 4.88 5.92 5.92 5.92 5.92
238 51 (41 to 62) 49
replacement (1.56-6.72) (1.56-6.72) (2.50-9.40) (3.10-11.14) (3.10-11.14) (3.10-11.14) (3.10-11.14)
8.33 8.33 8.33 8.33 8.33 8.33
Lateral resurfacing 13 57 (52 to 61) 46
(1.22-46.10) (1.22-46.10) (1.22-46.10) (1.22-46.10) (1.22-46.10) (1.22-46.10)
Distal humeral 3.70
Elective
27 71 (67 to 81) 30
hemiarthroplasty (0.53-23.51)
Unconfirmed total 2.31 3.77 6.92 9.15 9.15 10.73 11.82 11.82
220 67.5 (57.5 to 75) 30
elbow replacement (0.97-5.46) (1.90-7.39) (4.15-11.43) (5.86-14.16) (5.86-14.16) (7.00-16.26) (7.73-17.84) (7.73-17.84)
Unconfirmed radial
10 60 (48 to 77) 40
head replacement
Unconfirmed lateral
10 59.5 (45 to 68) 40
resurfacing
Unconfirmed
distal humeral 4 75.5 (66 to 80.5) 0
hemiarthroplasty
Note: Elbow replacements with a mismatch between the type of procedure reported by the surgeon on the MDS form and the recorded component labels on the MDS form or with no component data in the record
are described as unconfirmed and classified according to the procedure type indicated by the surgeon on the MDS form.
Table 3.E6 shows Kaplan-Meier estimates of the primary elbow replacement, divided into acute trauma
cumulative percentage probability of revision up to and elective cases. It should be noted that there are
eight years after the primary operation, together with substantial differences in the proportions of different
95% confidence intervals for all cases and for acute types of elbow replacement in the elective and trauma
trauma and elective cases separately. group that are likely to account for the differences
observed. Total elbow replacement makes up a higher
There is a higher cumulative revision rate for all elbow proportion of procedures in elective cases (82.0%)
arthroplasty for elective indications compared to than trauma (31.6%), whereas isolated radial head
trauma. Figure 3.E4 shows Kaplan-Meier estimates of replacement is more commonly performed in trauma
the cumulative percentage probability of revision after cases (49.1%) than elective (8.2%).
Figure 3.E4 KM estimates of cumulative revision of primary elbow replacement by acute trauma
and elective cases. Blue italics in the numbers at risk table signify that fewer than 250 cases remained
10

0
0 1 2 3 4 5 6 7 8
Years since primary
Acute trauma 2,135 1,687 1,203 848 607 393 225 131 38
Elective 2,908 2,637 2,178 1,733 1,302 926 626 376 130
Figure 3.E5 KM estimates of cumulative revision of primary total elbow replacement (with or
without a radial head replacement) by acute trauma and elective cases. Blue italics in the numbers
at risk table signify that fewer than 250 cases remained at risk at these time points.
10
8
0
0 1 2 3 4 5 6 7 8
Years since primary
Acute trauma 674 553 444 344 264 183 103 67 19
Elective 2,386 2,183 1,804 1,433 1,061 745 494 280 95
For the sub-group of total elbow replacement, shown in the registry and because the confidence intervals
in Figure 3.E5, we found that the survival of total of the estimates in both groups overlap. There is
replacements was comparable for trauma and elective insufficient data to compare lateral resurfacing, distal
indications up to two years. From two years post- humeral hemiarthroplasty and the other unconfirmed
operation onwards, the revision rates were higher for types of primary procedure between elective and
the elective total elbow replacements, but the data for trauma indications.
acute trauma is less certain due to the low numbers
Figure 3.E6 KM estimates of cumulative revision of primary radial head replacement by acute
trauma and elective cases. Blue italics in the numbers at risk table signify that fewer than 250 cases
10

0
0 1 2 3 4 5 6 7 8
Years since primary
Acute trauma 1,048 822 560 379 255 156 85 38 13
Elective 238 196 142 103 64 37 19 13 8
Figure 3.E6 shows Kaplan-Meier estimates of the the Department of Health and Social Care. For the
cumulative percentage probability of revision by purposes of the annual report, revision procedures
acute trauma and the elective cases in radial head include any addition, removal or modification of the
replacements. Revision of radial head replacement implants and procedures such as debridement and
appears to be under-reported as they are frequently implant retention with or without implant exchange,
revised to an excision arthroplasty which is often excision arthroplasty, amputation and conversion
poorly recorded by units. to arthrodesis. For the avoidance of confusion,
completing a revision MDS form is also mandatory
The NJR asks surgeons and those responsible for for a procedure involving modification of a joint by
healthcare delivery to ensure that when primary and adding another implant to another part of the joint.
revision joint replacement procedures of the hip, For the analyses of surgeon performance, hospital
knee, ankle, elbow or shoulder are performed, that performance and implant performance, debridement
the relevant MDS form is completed and data entered and implant retention without implant exchange is
into the registry. This is a requirement mandated by currently excluded.
Figure 3.E7 KM estimates of cumulative revision of total elbow replacements and distal humeral
hemiarthroplasty within the acute trauma cases. Blue italics in the numbers at risk table signify that
10
8
0
0 1 2 3 4 5 6 7 8
Years since primary
Distal humeral hemiarthroplasty 203 125 38 <4 <4 <4
Total elbow replacement 673 552 444 344 264 183 103 67 19
Figure 3.E7 shows cumulative rates of revision within

the acute trauma cases. These differences remain
uncertain as the number of procedures and the
number of revisions within these groups remain low
and excisions of radial head replacements are likely to
have been under-reported.
There are too few cases for further sub-division into

age/gender sub-groups.
Figure 3.E8 KM estimates of cumulative revision of total elbow replacements by implant brand within
the elective cases. Blue italics in the numbers at risk table signify that fewer than 250 cases remained at
risk at these time points (elbow replacements with less than 100 procedures are excluded).
15
12

0
0 1 2 3 4 5 6 7 8
Years since primary operation
Nexel 170 143 109 74 39 10
Latitude EV Stem 184 155 95 45 19 5 <4
Discovery 615 570 506 425 307 230 157 86 26
Coonrad Morrey 1,182 1,095 917 747 589 431 288 162 57
Table 3.E7 overleaf shows the cumulative probability Figure 3.E8 shows the rate of revision by implant
of revision for brands used in at least 100 primary brand within the elective cases. Brand comparisons
elbow replacements with a confirmed procedure type. will become more reliable as the size of the elbow
For total elbow replacement, the cumulative revision cohort increases over time, and allow further
rates varied between brands from 0.5% to 2.2% in the stratification by patient characteristics, acute/elective
first post-operative year. At five years post-operation, status and indication for primary surgery.
the rates still varied between brands from 5.9% to
9.0%. However, we note that as numbers are small,
this may simply be due to chance. For radial head
replacement, the cumulative revision rates varied
between brands from 0.4% to 2.1% in the first post-
operative year.
250
Table 3.E7 KM estimates of cumulative revision (95% CI) for all primary elbow procedures by implant brand. Blue italics signify that fewer than 250 cases
Time since primary

Number
of Age Male
primaries (Median,IQR) (%) 1 year 2 years 3 years 4 years 5 years 6 years 7 years 8 years
Coonrad 1.39 3.40 4.71 5.36 5.90 7.07 7.35 7.35
1,563 72 (64 to 78) 26
Morrey (0.91-2.13) (2.57-4.48) (3.69-6.00) (4.24-6.76) (4.69-7.41) (5.61-8.90) (5.82-9.27) (5.82-9.27)
0.54 1.41 3.18 5.92 7.14 8.26 8.79 9.96
Discovery 777 70 (61 to 78) 28
Total elbow Linked (0.20-1.42) (0.76-2.61) (2.06-4.91) (4.22-8.27) (5.16-9.82) (6.02-11.28) (6.39-12.04) (6.94-14.20)
replacement brands Latitude 1.19 1.90 3.86 7.07
179 71 (63 to 77) 24
EV Stem (0.30-4.67) (0.61-5.80) (1.24-11.67) (2.42-19.70)
2.21 3.98 3.98 6.92 9.04
Nexel 240 71 (62 to 79) 28
(0.92-5.23) (1.99-7.88) (1.99-7.88) (3.37-13.95) (4.44-17.92)
0.75 0.95 1.34 1.34 1.34 1.34 1.34 1.34
Anatomic 723 52 (41 to 63) 44
(0.31-1.81) (0.42-2.11) (0.60-2.97) (0.60-2.97) (0.60-2.97) (0.60-2.97) (0.60-2.97) (0.60-2.97)

Mono-
Radial head Evolve 0.37 0.37 0.37 0.37 0.37 0.37
polar 276 54 (41.5 to 64) 44
replacement (0.05-2.58) (0.05-2.58) (0.05-2.58) (0.05-2.58) (0.05-2.58) (0.05-2.58)
brands Proline
2.12 2.12 2.12 2.12 2.12
ExploR 104 50.5 (40 to 61.5) 38
(0.53-8.20) (0.53-8.20) (0.53-8.20) (0.53-8.20) (0.53-8.20)
Distal humeral Latitude 4.28 9.03
213 71 (65 to 79) 21
hemiarthroplasty EV Stem (2.15-8.43) (4.78-16.70)
Note: Elbow replacements with less than 100 procedures are excluded from this table.
Table 3.E8 gives a breakdown of the indications for revisions more than one indication was stated. A few
the first data-linked revision procedure. The most cases (n=56) had gone on to have further revision
common indications for revision remain aseptic procedures. The numbers are too small for any further
loosening and infection. The indications for revision analysis nor to draw any reliable conclusions.
were not mutually exclusive; in 26 of the 219 first
Table 3.E8 Indications for first data linked revision after any primary elbow replacement. Acute trauma and elective
cases are shown separately, for total elbow replacement, lateral resurfacing and distal humeral hemiarthroplasty, and
radial head replacement.
Indication for first revision procedure
Other
Number indication Peripros-
of Total Aseptic Failed hemi- for thetic
Type of primary procedure primaries revised loosening arthroplasty Infection Instability revision fracture
All acute trauma and
5,043 219 85 11 73 29 17 32
elective cases
Confirmed elbow
1,925 41 13 <4 14 7 5 4
replacements
Total elbow

673 25 11 0 11 <4 <4 4
replacement
Total elbow
replacement inc. radial <4 0 0 0 0 0 0 0
head replacement
Radial head
1,048 6 <4 0 <4 <4 <4 0
replacement
Lateral resurfacing 0 0 0 0 0 0 0 0
Acute trauma
Distal humeral
203 10 0 <4 <4 5 <4 0
hemiarthroplasty
Unconfirmed elbow
210 7 <4 <4 <4 <4 <4 0
replacements
Unconfirmed total
163 <4 <4 <4 0 <4 <4 0
elbow replacement
Unconfirmed radial
37 <4 <4 0 0 <4 0 0
head replacement
Unconfirmed lateral
<4 0 0 0 0 0 0 0
resurfacing
Unconfirmed
distal humeral 8 <4 0 <4 <4 <4 0 0
hemiarthroplasty
Note: Elbow replacements with a mismatch between the type of procedure reported by the surgeon on the MDS form and the recorded component labels on the MDS
form or with no component data in the record are described as unconfirmed and classified according to the procedure type indicated by the surgeon on the MDS form.
Indication for first revision procedure
Other
Number indication Peripros-
of Total Aseptic Failed hemi- for thetic
Type of primary procedure primaries revised loosening arthroplasty Infection Instability revision fracture
Confirmed elbow
2,664 148 59 4 50 15 10 27
replacements
Total elbow
2,358 133 57 0 48 11 7 26
replacement
Total elbow
replacement inc. radial 28 <4 0 0 0 0 <4 <4
head replacement
Radial head
238 10 <4 <4 <4 <4 <4 0
replacement
Lateral resurfacing 13 <4 0 0 0 0 <4 0
Elective
Distal humeral
27 <4 0 <4 0 <4 0 0
hemiarthroplasty
Unconfirmed elbow
244 23 11 <4 8 4 <4 <4
replacements
Unconfirmed total
220 21 11 <4 8 <4 <4 0
elbow replacement
Unconfirmed radial
10 0 0 0 0 0 0 0
head replacement
Unconfirmed lateral
10 <4 0 0 0 <4 0 <4
resurfacing
Unconfirmed
distal humeral 4 0 0 0 0 0 0 0
hemiarthroplasty
Note: Elbow replacements with a mismatch between the type of procedure reported by the surgeon on the MDS form and the recorded component labels on the MDS
form or with no component data in the record are described as unconfirmed and classified according to the procedure type indicated by the surgeon on the MDS form.
3.5.3 Mortality after primary elbow

replacement surgery
For this analysis, the second procedure of a pair of
bilateral operations performed on the same day were
excluded (Figure 3.E1 on page 231). Among the
remaining 5,033 procedures, 606 of the recipients had
died by the end of December 2020.
Table 3.E9 KM estimates of cumulative mortality (95% CI) by time from primary elbow replacement, for acute trauma and elective cases.
Number Age Time since primary

of (Median, Male
primaries IQR) (%) 30 days 90 days 1 year 2 years 3 years 4 years 5 years 6 years 7 years 8 years
All acute trauma and 0.22 0.58 2.24 4.83 8.22 11.57 16.03 19.93 23.09 27.16
5,033 67 (56 to 76) 31
elective cases (0.12-0.40) (0.41-0.84) (1.86-2.70) (4.23-5.51) (7.39-9.14) (10.52-12.71) (14.71-17.47) (18.33-21.65) (21.22-25.09) (24.68-29.85)
All acute 0.43 0.81 2.93 5.99 10.10 13.69 17.85 22.46 25.81 29.33
2,128 66 (52 to 77) 30
trauma cases (0.22-0.82) (0.51-1.30) (2.27-3.79) (4.96-7.23) (8.63-11.82) (11.83-15.81) (15.52-20.48) (19.50-25.79) (22.24-29.84) (24.54-34.81)
Total elbow 0.75 1.66 5.97 11.27 17.47 23.22 28.53 34.49 37.39 39.87
672 77 (71 to 83) 17
replacement (0.31-1.79) (0.92-2.97) (4.37-8.11) (8.97-14.12) (14.50-20.97) (19.71-27.25) (24.50-33.05) (29.73-39.78) (32.10-43.23) (33.87-46.51)
Total elbow
replacement
<4 79 (79 to 79) 0
inc. radial head
replacement
Radial head 0.19 0.29 0.75 1.42 1.86 2.50 2.99 4.59 9.53 9.53
1,042 53 (41 to 63) 42
replacement (0.05-0.78) (0.09-0.90) (0.36-1.57) (0.81-2.51) (1.08-3.20) (1.46-4.27) (1.72-5.16) (2.51-8.33) (4.87-18.19) (4.87-18.19)
Distal humeral 0.53 3.21 7.97
203 71 (65 to 79) 18 0
hemiarthroplasty (0.07-3.70) (1.35-7.58) (4.28-14.59)
Unconfirmed
1.23 1.23 4.31 9.40 18.83 23.63 32.85 38.35 40.55
total elbow 163 75 (66 to 83) 20
Acute trauma
(0.31-4.82) (0.31-4.82) (2.08-8.84) (5.78-15.12) (13.47-25.98) (17.36-31.69) (24.75-42.76) (29.06-49.40) (30.76-52.08)
replacement
Unconfirmed
radial head 37 48 (37 to 59) 51 0 0 0 0 0
replacement
Unconfirmed
lateral <4 74.5 (74 to 75) 50
resurfacing
Unconfirmed
distal humeral 8 74 (63 to 80.5) 38
hemiarthroplasty
Note: Elbow replacements with a mismatch between the type of procedure reported by the surgeon on the MDS form and the recorded component labels on the MDS form or with no component data in the record are described as unconfirmed and classified
according to the procedure type indicated by the surgeon on the MDS form.
253
254
Number Age Time since primary

of (Median, Male
primaries IQR) (%) 30 days 90 days 1 year 2 years 3 years 4 years 5 years 6 years 7 years 8 years
All elective 0.07 0.42 1.77 4.07 7.10 10.34 14.96 18.61 21.72 25.98
2,905 68 (58 to 75) 31
cases (0.02-0.28) (0.24-0.73) (1.35-2.33) (3.38-4.90) (6.14-8.20) (9.12-11.71) (13.38-16.70) (16.75-20.66) (19.56-24.09) (23.12-29.12)
Total elbow 0.08 0.43 1.87 4.53 7.91 11.55 16.59 20.62 23.76 27.58
2,356 69 (60 to 76) 29
replacement (0.02-0.34) (0.23-0.79) (1.39-2.51) (3.73-5.49) (6.79-9.19) (10.14-13.15) (14.78-18.61) (18.47-22.98) (21.30-26.46) (24.33-31.18)
Total elbow
replacement
28 64 (55 to 71) 36 0 0 0
inc. radial head
replacement
Radial head 0.49 0.49 0.49 2.85 2.85 2.85 2.85
237 51 (41 to 62) 49 0 0
replacement (0.07-3.44) (0.07-3.44) (0.07-3.44) (0.85-9.36) (0.85-9.36) (0.85-9.36) (0.85-9.36)
Lateral
13 57 (52 to 61) 46 0 0 0 0 0 0 0
resurfacing
Distal humeral
27 71 (67 to 81) 30 0 0 0
hemiarthroplasty
Elective
Unconfirmed
0.91 2.73 4.13 6.10 7.17 11.18 13.96 16.86 24.52
total elbow 220 67.5 (57.5 to 75) 30
(0.23-3.59) (1.24-5.98) (2.17-7.78) (3.59-10.29) (4.38-11.63) (7.48-16.53) (9.66-19.96) (11.83-23.72) (17.50-33.74)

replacement
Unconfirmed
radial head 10 60 (48 to 77) 40
replacement
Unconfirmed
lateral 10 59.5 (45 to 68) 40
resurfacing
Unconfirmed
distal humeral 4 75.5 (66 to 80.5) 0
hemiarthroplasty
Note: Elbow replacements with a mismatch between the type of procedure reported by the surgeon on the MDS form and the recorded component labels on the MDS form or with no component data in the record are described as unconfirmed and classified
according to the procedure type indicated by the surgeon on the MDS form.
Figure 3.E9 KM estimates of cumulative mortality of primary elbow replacement by acute trauma
and elective cases. Blue italics in the numbers at risk table signify that fewer than 250 cases remained
35
30

25
20
15
10
0
0 1 2 3 4 5 6 7 8
Years since primary
Acute trauma 2,128 1,702 1,222 865 623 404 231 133 38
Elective 2,905 2,674 2,246 1,819 1,394 1,004 689 417 143
Table 3.E9 and Figure 3.E9 show the overall The mortality rate at four years after primary total
cumulative percentage probability of mortality shown elbow replacement for trauma is 101.0% higher than
separately for acute trauma and the elective cases. the rate in elective total elbow arthroplasty, with a four-
year mortality rate of 23.2%.
3.5.4 Conclusions years of data entry with inflammatory arthropathy
accounting for 41.7% of cases. In 2020 there were
The annual number of primary elbow replacement 215 confirmed elective and acute trauma primary
procedures entered into the registry has increased total elbow replacements (including three with radial
since 2012, other than in 2020 which was profoundly head replacements) performed in 88 units by 101
affected by COVID-19. The NJR has one of the consultants. The volume of procedures does not
largest datasets of elbow arthroplasty globally. It is show large variation, however the number of units
not yet known how accurate or complete the dataset performing elbow replacements has declined, from
is, and an independent audit of elbow replacement 125 in 2018, and the number of consultants from 151
data is underway. in 2018. It has been the intention of the NHSE/I GIRFT
programme to centralise total elbow replacement
The type of procedure reported is determined from surgery across fewer specialist centres, so this data
two sources of information. The first is the procedure is encouraging, although this comparison may have
type recorded on the MDS data collection form by the been affected by the impact of COVID-19 on the 2020
surgeon, or their deputy, at the time of the procedure. figures. It should be noted that the median numbers
The second source is the set of component labels of primary procedures per unit and per surgeon have
attached to the MDS form and recorded at upload of not changed significantly from 2018 to those reported
the record. When there is a mismatch between these in 2020.
two sources, i.e. the components entered do not match
the procedure type recorded or in the case where The Kaplan-Meier estimate of cumulative revision of
there is no component data at all in the data entry total elbow replacement at four years was 4.29 (95%
record, the procedure type is reported as unconfirmed. CI 2.77-6.60) for trauma patients and 6.34 (95% CI
Further work is required to reconcile these unconfirmed 5.28-7.61) for elective cases. Disparities in the rate
procedures and reduce their ‘unconfirmed’ status. This of revision were observed between implant brands.
will enhance the comprehensiveness and utility of the Brand comparisons will become more evident and
data moving forward. reliable as the size of the elbow cohort increases over
time. We note that the main indications for revision
Distal humeral hemiarthroplasty was not included were infection and aseptic loosening and this is
in the MDS until June 2018. Despite this, their use observed for both acute trauma and elective cases.
appears to be increasing overall, but total numbers
remain low, so it is not yet possible to compare the The 5-year mortality rate for elbow replacement in all
revision rates for this newer procedure against the cases is 16.03 (95% CI 14.71-17.47) with differences
data for total elbow replacement. Most distal humeral seen between trauma and elective surgery. The
hemiarthroplasty and radial head replacement 1-year mortality rate following total elbow replacement
procedures are performed for acute trauma and remains higher in the trauma patient population than
trauma sequelae, as expected. in those having elective surgery, however this is likely
to represent a difference in the demographics of these
The distribution of indications for elective total elbow two patient groups.
replacement has been consistent over the five
3.6 Outcomes
after shoulder
replacement
3.6.1 Overview of primary shoulder to build a construct are present in a procedure. We
have cross-checked the implanted construct with
replacement surgery the indicated procedure at the time of the surgery
Shoulder replacements have been recorded in the and positively confirmed the implanted construct
registry since 2012. In this section we address matches the reported procedure. This has led to the
an overview of the (data-linked) primary shoulder definition of unconfirmed constructs of which there
replacements performed up to 31 December 2020 are either insufficient implants listed to make up a
and also document the first revision and mortality, complete construct, or the implants used do not
when these events had occurred following a primary match the indicated procedure. A total of 4,774 (9.5%)
shoulder replacement. procedures are unconfirmed; although the volume
is expected to improve in future reports, with the
In 2018 and 2019 a rigorous review of the shoulder development of more rigorous checks.
data was undertaken due to the rapid expansion of
shoulder implant types available. As a consequence We define a stemmed humeral component as
of this review, new classifications and component a humeral component in which any part enters
attributes are now used within the report to define the humeral diaphysis, while a stemless humeral
the primary groupings throughout the whole of this component is defined as being completely confined to
section. The report has now moved to whole construct the metaphysis with no part entering the diaphysis.
validation, ensuring all relevant elements required
National Joint Registry | 18th Annual Report | Shoulders
Figure 3.S1 Shoulder cohort flow diagram.
Shoulder procedures recorded by the NJR

N=58,581
Non−consenting procedures N=1,590
Invalid IDs N=11

N=56,697

N=56,646

N=56,646
Isle of Man N=0

N=56,131
/ Prostheses used / Indications / Unit N=47
Unique procedures
N=56,082
Procedures (98 shoulders) with
Procedures (53,235 shoulders)

N=55,871

*Of which, shoulders procedures where first recorded
procedure is a revision N=3,528
Primary procedures
(Revision analyses)
N=50,255
Bilateral procedures (same day) N=25
N=50,230
A total of 50,255 primary shoulder replacements were bilateral simultaneous operations (left and right
were available for our analysis in a total of 46,277 on the same day). See Figure 3.S1 for a detailed
patients. Of these patients, 3,978 had documented description of patients included in this section.
replacements on both left and right sides, 25 of which
Table 3.S1 Number and percentage of primary shoulder replacements (elective or acute trauma), by year and
type of shoulder replacement.
All Year of primary

years 2012 2013 2014 2015 2016 2017 2018 2019 2020
N (%) N (%) N (%) N (%) N (%) N (%) N (%) N (%) N (%) N (%)
50,255 2,545 4,412 5,309 5,734 6,537 7,002 7,223 7,660 3,833
All cases
(100.0) (100.0) (100.0) (100.0) (100.0) (100.0) (100.0) (100.0) (100.0) (100.0)
Proximal humeral 8,064 885 1,303 1,287 1,060 1,015 836 707 679 292
hemiarthroplasty (16.0) (34.8) (29.5) (24.2) (18.5) (15.5) (11.9) (9.8) (8.9) (7.6)
2,906 476 594 537 375 368 220 146 130 60
Resurfacing
(5.8) (18.7) (13.5) (10.1) (6.5) (5.6) (3.1) (2.0) (1.7) (1.6)
1,227 70 132 164 139 164 171 175 167 45
Stemless
(2.4) (2.8) (3.0) (3.1) (2.4) (2.5) (2.4) (2.4) (2.2) (1.2)
3,931 339 577 586 546 483 445 386 382 187
Stemmed
(7.8) (13.3) (13.1) (11.0) (9.5) (7.4) (6.4) (5.3) (5.0) (4.9)
Total shoulder 13,734 632 1,178 1,534 1,770 1,897 1,983 1,892 1,933 915
replacement (27.3) (24.8) (26.7) (28.9) (30.9) (29.0) (28.3) (26.2) (25.2) (23.9)
486 49 99 82 88 78 45 24 15 6
Resurfacing
(1.0) (1.9) (2.2) (1.5) (1.5) (1.2) (0.6) (0.3) (0.2) (0.2)
4,915 137 256 390 505 631 733 855 937 471
Stemless
(9.8) (5.4) (5.8) (7.3) (8.8) (9.7) (10.5) (11.8) (12.2) (12.3)
8,333 446 823 1,062 1,177 1,188 1,205 1,013 981 438
Stemmed
(16.6) (17.5) (18.7) (20.0) (20.5) (18.2) (17.2) (14.0) (12.8) (11.4)
Reverse polarity
23,678 686 1,351 1,907 2,328 3,008 3,609 3,969 4,512 2,308
total shoulder
(47.1) (27.0) (30.6) (35.9) (40.6) (46.0) (51.5) (54.9) (58.9) (60.2)
replacement
186 5 14 15 25 25 21 38 23 20
Stemless
(0.4) (0.2) (0.3) (0.3) (0.4) (0.4) (0.3) (0.5) (0.3) (0.5)
23,492 681 1,337 1,892 2,303 2,983 3,588 3,931 4,489 2,288
Stemmed
(46.7) (26.8) (30.3) (35.6) (40.2) (45.6) (51.2) (54.4) (58.6) (59.7)
Interpositional 5 0 0 0 0 0 0 <4 <4 0
arthroplasty (<0.1) (0) (0) (0) (0) (0) (0) (<0.1) (<0.1) (0)
4,774 342 580 581 576 617 574 653 533 318
Unconfirmed
(9.5) (13.4) (13.1) (10.9) (10.0) (9.4) (8.2) (9.0) (7.0) (8.3)
346 21 59 40 42 40 34 45 42 23
Unconfirmed HHA
(0.7) (0.8) (1.3) (0.8) (0.7) (0.6) (0.5) (0.6) (0.5) (0.6)
1,853 201 312 304 258 271 205 166 79 57
Unconfirmed TSR
(3.7) (7.9) (7.1) (5.7) (4.5) (4.1) (2.9) (2.3) (1.0) (1.5)
2,570 120 209 237 276 306 335 438 411 238
Unconfirmed RTSR
(5.1) (4.7) (4.7) (4.5) (4.8) (4.7) (4.8) (6.1) (5.4) (6.2)
5 0 0 0 0 0 0 4 <4 0
Unconfirmed IPA
(<0.1) (0) (0) (0) (0) (0) (0) (0.1) (<0.1) (0)
Note: HHA=Proximal humeral hemiarthroplasty, TSR=Total shoulder replacement, RTSR=Reverse polarity total shoulder replacement, IPA=Interpositional arthroplasty.
Table 3.S1 illustrates the number of shoulder elective and acute trauma settings i.e. if a surgeon
replacements and how they have changed across performed 24 elective primary stemmed humeral
time. There is a steady increase in the number of hemiarthroplasty procedures and 24 acute stemmed
primary shoulder replacements year-on-year. It also humeral hemiarthroplasty procedures their annual
illustrates relative proportions of proximal humeral total volume would be 48 procedures. Those 48
hemiarthroplasty (HHA), conventional total shoulder procedures would contribute to the grey sub-division
replacement (TSR) and reverse polarity total shoulder in both elective and acute trauma figures.
replacement (RTSR). There is a continued increasing
preference for reverse polarity total shoulder Figure 3.S2 shows a complex pattern of increasing
replacement year-on-year. and decreasing treatment preferences. Resurfacing
humeral hemiarthroplasty and total shoulder
The number of unconfirmed procedures contained replacements have declined since the start of data
within the registry is illustrated. Using more evolved collection, stemless total shoulder replacements have
methods of construct and procedure cross-validation, steadily increased, the volume of stemmed reverse
procedures with insufficient prostheses elements to polarity total shoulder replacement has increased
build a unique construct or a construct that disagrees substantially, and stemmed humeral hemiarthroplasty
with the procedure indicated at the time of surgery are and total shoulder replacements have fallen. In the
identified. It is noted that entering all the elements of more common procedures (stemless total shoulder
reverse polarity total shoulder replacements appears to replacements, stemmed total shoulder replacements
be particularly challenging and so it is urged that those and stemmed reverse polarity total shoulder
completing the data entry forms and entering data replacements), the growth in procedures appears to
should pay particular attention to these procedures. be occurring in higher volume shoulder surgeons.
Figure 3.S2 and Figure 3.S3 overleaf show the yearly Figure 3.S3 shows the popularity of stemmed humeral
number of primary shoulder replacements performed hemiarthroplasty has reduced over the last few years
for elective and acute trauma indications respectively. while the popularity of stemmed reverse polarity total
Elective and acute trauma procedures have been shoulder replacements has been steadily increasing.
stratified by procedure type. Please note the difference Stemmed reverse polarity total shoulder replacements
in scale of the y-axis between each sub-plot. Each are increasingly conducted by higher volume surgeons.
bar is further stratified by the volume of procedures
that the surgeon conducted in that year across both
Figure 3.S2 Frequency of primary shoulder replacements within elective patients stratified by
procedure type, bars stacked by volume per consultant per year.
Resurfacing HHA Stemless HHA Stemmed HHA
500 150 250
400 200
100
300 150
200 100
50
100 50
0 0 0
12 14 16 18 20 12 14 16 18 20 12 14 16 18 20
13 15 17 19 13 15 17 19 13 15 17 19
Resurfacing TSR Stemless TSR Stemmed TSR

80 800 1,000
Frequency (N=)
800
60 600
600
40 400
400
20 200
200
0 0 0
12 14 16 18 20 12 14 16 18 20 12 14 16 18 20
13 15 17 19 13 15 17 19 13 15 17 19
Stemless RTSR Stemmed RTSR
4,000
30
3,000
20
2,000
10
1,000
0 0
12 14 16 18 20 12 14 16 18 20
13 15 17 19 13 15 17 19
1≤N≤2 3≤N≤4 5≤N≤6 7≤N≤12 13≤N≤24 25≤N≤48 49≤N≤96
Figure 3.S3 Frequency of primary shoulder replacements within acute trauma patients stratified by
procedure type, bars stacked by volume per consultant per year.
Stemmed HHA Stemmed RTSR
150 600

Frequency (N=)
100 400
50 200
0 0
12 14 16 18 20 12 14 16 18 20
13 15 17 19 13 15 17 19
1≤N≤2 3≤N≤4 5≤N≤6 7≤N≤12 13≤N≤24 25≤N≤48 49≤N≤96
Figure 3.S4 Age (Box and whiskers*) and frequency of primary shoulder replacements by gender and
type of shoulder replacement.
Proximal humeral hemiarthroplasty

Resurfacing
Stemless
Stemmed
Total shoulder replacement

Resurfacing
Stemless
Stemmed
Reverse polarity total shoulder replacement

Stemless
Stemmed
Interpositional arthroplasty
Unconfirmed
Unconfirmed HHA
Unconfirmed TSR
Unconfirmed RTSR
Unconfirmed IPA
20
30
40
50
60
70
80
90
30
60
90
0
10
12
15
Age in years Frequency x100
Female Male
*"Box and Whiskers" | represents the median, boxes represent lower and upper interquartile range.
Whiskers represent the 2.5th and 97.5th centile of the distribution.
Figure 3.S4 illustrates the age and gender difference total shoulder replacements tend to be older than
between the different types and sub-types of those receiving proximal humeral hemiarthroplasty
shoulder replacements using a modified ‘box and or conventional total shoulder replacements. Figure
whisker’ plot. The whiskers represent the 2.5th 3.S4 also illustrates that the majority of procedures
and 97.5th centile of the distribution. The figure recorded within the registry are reverse polarity total
also shows the frequency of procedures by gender shoulder replacements. It also clearly illustrates that
and procedure type. The plots illustrate the points the majority of unconfirmed procedures consist of
that women tend to be older than men at the time reverse polarity total shoulder replacements.
of operation and those receiving reverse polarity
Table 3.S2 Demographic characteristics of patients undergoing primary shoulder replacements, by acute or
elective indications and type of shoulder replacement.
Number of Male Age in years at primary
Shoulder type cases N (%) median (IQR*) range**
All cases 5,143 1,180 (22.9) 74 (67 to 80) 27 to 99
Acute trauma
Proximal humeral hemiarthroplasty 1,648 499 (30.3) 68 (60 to 77) 27 to 96

Total shoulder replacement 14 8 (57.1) 69 (53 to 74) 43 to 79
Reverse polarity total shoulder replacement 3,011 577 (19.2) 76 (70 to 81) 48 to 99
Interpositional arthroplasty 0 0 (0.0) 0 (0 to 0) 0 to 0
Unconfirmed 470 96 (20.4) 74 (68 to 80) 35 to 95
All cases 45,112 13,761 (30.5) 73 (67 to 79) 17 to 100
Proximal humeral hemiarthroplasty 6,416 2,137 (33.3) 70 (61 to 77) 17 to 95

Resurfacing 2,900 897 (30.9) 71 (64 to 78) 20 to 95
Stemless 1,217 514 (42.2) 67 (56 to 75) 17 to 93
Stemmed 2,299 726 (31.6) 70 (60 to 78) 19 to 95
Total shoulder replacement 13,720 4,319 (31.5) 70 (64 to 76) 18 to 99
Resurfacing 486 140 (28.8) 71 (63 to 76) 29 to 95
Stemless 4,911 1,763 (35.9) 69 (62 to 75) 18 to 99
Elective
Stemmed 8,323 2,416 (29.0) 71 (65 to 76) 24 to 96

Reverse polarity total shoulder replacement 20,667 5,902 (28.6) 76 (71 to 80) 17 to 100
Stemless 186 70 (37.6) 73 (69 to 78) 49 to 89
Stemmed 20,481 5,832 (28.5) 76 (71 to 80) 17 to 100
Interpositional arthroplasty 5 <4 (60.0) 58 (55 to 68) 42 to 73
Unconfirmed 4,304 1,400 (32.5) 73 (66 to 78) 18 to 96
Unconfirmed HHA 292 109 (37.3) 69 (59 to 76) 18 to 92
Unconfirmed TSR 1,814 644 (35.5) 69 (61 to 76) 20 to 96
Unconfirmed RTSR 2,195 644 (29.3) 75 (69 to 80) 18 to 95
Unconfirmed IPA <4 <4 (100.0) 60 (58 to 65) 58 to 65
*IQR: Interquartile range, i.e. 25th and 75th centile.

**Range: Lowest and highest observed values.
Table 3.S2 displays similar information to Figure 3.S4,

except results are divided by acute trauma and
elective procedures.
Table 3.S3 Numbers of units and consultant surgeons providing primary shoulder replacements and median and
interquartile range of procedures performed by unit and consultant, by year, last five years and overall.
Units providing Consultants
primary Primary providing primary Primary
Primary replacements in replacements replacements in replacements
replacements each year per unit each year per consultant
Year of primary N N Median (IQR) N Median (IQR)
All years 50,255 409 80 (28 to 172) 880 19 (2 to 85)
Last 5 years 32,255 399 56 (21 to 114) 700 28 (4 to 72.5)

2012 2,545 262 6 (3 to 12) 379 4 (2 to 9)
2013 4,412 312 9 (4 to 18) 432 7 (2 to 15)
2014 5,309 338 10 (4 to 21) 456 8 (3 to 17)
2015 5,734 347 11 (4 to 23) 486 8 (3 to 17)
2016 6,537 348 14 (5 to 26) 490 10 (4 to 19)
2017 7,002 364 14 (5 to 27) 492 11 (5 to 21)
2018 7,223 367 14 (5 to 28) 506 11 (4 to 21)
2019 7,660 374 14 (6 to 29) 518 11 (4 to 21)
2020 3,833 350 7 (3 to 15) 471 6 (3 to 12)
Table 3.S3 illustrates the number of primary shoulder by year of data collection. The results illustrate that the
replacements and the number of units and consultants median, and interquartile range, number of procedures
conducting shoulder replacements within the registry. performed by units and consultants has remained
The table also illustrates the median and interquartile static for the last few years, with the exception of 2020
range of the number of replacements performed within and COVID-19, it has now fallen to 7 (3 to 15) and
each unit or by each consultant. This is displayed 6 (3 to 12) procedures respectively.
overall, aggregated by the last five years of data, and
Table 3.S4 Number and percentage of primary shoulder replacements by indication and type of shoulder replacement.
Acute trauma Elective

N (%)* for each indication in elective procedures only
Number Number Other Cuff tear
of cases of cases Cuff tear Trauma inflamatory Avascular Other without
N (%) N (%) Osteoarthritis arthropathy sequelae arthropathy necrosis causes*** arthropathy**
All cases 5,143 (100.0) 45,112 (100.0) 27,297 (100.0) 12,584 (100.0) 3,182 (100.0) 1,819 (100.0) 1,447 (100.0) 1,046 (100.0) 584 (100.0)
Proximal humeral
1,648 (32.0) 6,416 (14.2) 4,809 (17.6) 353 (2.8) 557 (17.5) 392 (21.6) 521 (36.0) 181 (17.3) 7 (1.2)
hemiarthroplasty
Resurfacing 6 (0.1) 2,900 (6.4) 2,465 (9.0) 166 (1.3) 70 (2.2) 160 (8.8) 102 (7.0) 50 (4.8) <4 (0.3)
Stemless 10 (0.2) 1,217 (2.7) 984 (3.6) 18 (0.1) 81 (2.5) 61 (3.4) 122 (8.4) 36 (3.4) 0 (0)
Stemmed 1,632 (31.7) 2,299 (5.1) 1,360 (5.0) 169 (1.3) 406 (12.8) 171 (9.4) 297 (20.5) 95 (9.1) 5 (0.9)
Total shoulder
14 (0.3) 13,720 (30.4) 12,746 (46.7) 34 (0.3) 277 (8.7) 497 (27.3) 347 (24.0) 174 (16.6) 4 (0.7)
replacement
Resurfacing 0 (0) 486 (1.1) 464 (1.7) 0 (0) 4 (0.1) 22 (1.2) <4 (0.1) 4 (0.4) 0 (0)
Stemless 4 (0.1) 4,911 (10.9) 4,539 (16.6) 11 (0.1) 107 (3.4) 183 (10.1) 114 (7.9) 80 (7.6) <4 (0.3)
Stemmed 10 (0.2) 8,323 (18.4) 7,743 (28.4) 23 (0.2) 166 (5.2) 292 (16.1) 231 (16.0) 90 (8.6) <4 (0.3)
Reverse polarity total
3,011 (58.5) 20,667 (45.8) 7,346 (26.9) 10,961 (87.1) 1,978 (62.2) 723 (39.7) 452 (31.2) 474 (45.3) 531 (90.9)
shoulder replacement
Stemless 0 (0) 186 (0.4) 74 (0.3) 103 (0.8) 6 (0.2) <4 (0.1) 4 (0.3) 0 (0) 6 (1.0)
Stemmed 3,011 (58.5) 20,481 (45.4) 7,272 (26.6) 10,858 (86.3) 1,972 (62.0) 721 (39.6) 448 (31.0) 474 (45.3) 525 (89.9)
Interpositional
0 (0) 5 (<0.1) 5 (<0.1) 0 (0) 0 (0) 0 (0) 0 (0) 0 (0) 0 (0)
arthroplasty
Unconfirmed 470 (9.1) 4,304 (9.5) 2,391 (8.8) 1,236 (9.8) 370 (11.6) 207 (11.4) 127 (8.8) 217 (20.7) 42 (7.2)
Unconfirmed HHA 54 (1.0) 292 (0.6) 174 (0.6) 49 (0.4) 26 (0.8) 16 (0.9) 24 (1.7) 27 (2.6) 0 (0)
Unconfirmed TSR 39 (0.8) 1,814 (4.0) 1,459 (5.3) 127 (1.0) 77 (2.4) 79 (4.3) 48 (3.3) 92 (8.8) <4 (0.2)
Unconfirmed RTSR 375 (7.3) 2,195 (4.9) 755 (2.8) 1,060 (8.4) 267 (8.4) 112 (6.2) 55 (3.8) 98 (9.4) 41 (7.0)
Unconfirmed IPA <4 (<0.1) <4 (<0.1) <4 (<0.1) 0 (0) 0 (0) 0 (0) 0 (0) 0 (0) 0 (0)
*Percentages are based on the total number of elective cases; please note the listed reasons are not mutually exclusive as more than one reason could have been stated.
**Only recorded in MDSv7 introduced in June 2018. Total cases recorded using MDSv7 =16,280.
***Includes 80 metastatic cancer/malignancies documented since MDSv6 (N=40,241), together with 182 dislocations documented since MDSv7 (N=16,280).
267
Table 3.S4 illustrates the number and percentage of cuff tear arthropathy is the predominant indication
primary shoulder procedures by the type and sub- for reverse polarity total shoulder replacements. It
type of shoulder replacement for both acute trauma is important to note that the indications for surgery
and elective procedures. The indication for surgery in recorded in the registry are not mutually exclusive;
elective procedures is also illustrated. The majority of 84.7% of procedures list a single indication for the
proximal humeral hemiarthroplasty and conventional cause of surgery, with the remainder recording more
total shoulder replacement procedures recorded in the than one indication.
registry are for an indication of osteoarthritis, whereas
Table 3.S5 (a) Number of resurfacing proximal humeral hemiarthroplasty replacements between 2012 and
2020 and within the last year by brand construct.
Primary operations all years Primary operations in 2020
Acute Acute
All cases trauma Elective All cases trauma Elective

Manufacturer(s) Shoulder construct N N N N N N
Wright Aequalis Resurfacing[HH.Resurf] 251 0 251 0 0 0
FH Arrow[HH.Resurf] 35 0 35 0 0 0
Resurfacing HHA
Zimmer Biomet Copeland[HH.Resurf] 1,620 <4 1,617 30 0 30

DePuy Epoca[HH.Resurf] 112 <4 111 0 0 0
Exactech Equinoxe[HH.Resurf:H.RPeg] 42 0 42 6 0 6
DePuy Global CAP[HH.Resurf] 609 <4 607 14 0 14
Lima SMR[HH.Resurf:H.RPeg] 110 0 110 0 0 0
Lima SMR[HH.Resurf] 23 0 23 0 0 0
JRI Vaios[HH.Resurf] 100 0 100 10 0 10
Table 3.S5 (b) Number of stemless proximal humeral hemiarthroplasty replacements between 2012 and 2020
and within the last year by brand construct.
Acute Acute

Versa-Dial[HH.Stand]: Nano[H.
Zimmer Biomet 56 <4 55 <4 0 <4
Stemless]
Mathys Affinis[HH.Stand:H.Stemless] 586 5 581 26 <4 25
Stemless HHA
Arthrex Eclipse[HH.Stand:H.Stemless] 129 <4 128 <4 0 <4

Global ICON[HH.Stand:H.
DePuy 19 0 19 <4 0 <4
Stemless]
Lima SMR[HH.Stand:H.Stemless] 30 0 30 4 0 4
Zimmer Biomet Sidus[HH.Stand:H.Stemless] 170 <4 169 6 0 6
Wright Simpliciti[HH.Stand:H.Stemless] 159 0 159 <4 0 <4
Zimmer Biomet TESS[HH.Stand:H.Stemless] 75 <4 73 0 0 0
Table 3.S5 (c) Number of stemmed proximal humeral hemiarthroplasty replacements between 2012 and 2020
and within the last year by brand construct.
All Acute All Acute

cases trauma Elective cases trauma Elective
Aequalis[HH.Stand]: Aequalis-
Wright 217 184 33 14 11 <4
Fracture[H.Standard]
Bigliani/Flatow[HH.Stand]:
Zimmer Biomet 22 <4 19 0 0 0
Anatomical[H.Mod]
Aequalis[HH.Stand]: Ascend Flex[H.
Wright 251 6 245 27 0 27
Standard]
Versa-Dial[HH.Stand]:
Zimmer Biomet 143 8 135 6 0 6
Comprehensive[H.Standard]
Bio-Modular[HH.Stand]:
Comprehensive Fracture[H.Standard]
Versa-Dial[HH.Stand]:
Zimmer Biomet 184 146 38 19 16 <4
Comprehensive Fracture[H.Standard]
Global Unite[HH.Stand]: Global AP[H.
DePuy 10 0 10 <4 0 <4
Mod]
Global Advantage[HH.Stand]: Global
DePuy 210 169 41 <4 <4 0
FX[H.Standard]

Zimmer Biomet Bigliani/Flatow[HH.Stand]: TM[H.Dia] 24 <4 23 0 0 0
Wright Aequalis[HH.Stand:H.Standard] 196 4 192 0 0 0
Mathys Affinis[HH.Stand:H.NeckBody:H.Dia] 208 175 33 13 12 <4
Mathys Affinis[HH.Stand:H.Standard] 66 <4 63 <4 <4 <4
Stemmed HHA
Zimmer Biomet Anatomical[HH.Stand:H.Mod] 22 <4 20 <4 0 <4

Anatomical Fracture[HH.Stand:H.
Mod]
FH Arrow[HH.Stand:H.Standard] 33 5 28 <4 <4 0
Wright Ascend Flex[HH.Stand:H.Standard] 162 5 157 11 <4 10
Zimmer Biomet Bigliani/Flatow[HH.Stand:H.Dia] 47 12 35 0 0 0
Zimmer Biomet Bio-Modular[HH.Stand:H.Standard] 11 6 5 0 0 0
DePuy Delta Xtend[HH.Stand:H.Standard] 41 <4 39 0 0 0
DePuy Epoca[HH.Stand:H.Mod] 115 51 64 0 0 0
Exactech Equinoxe[HH.Stand:H.Standard] 223 191 32 25 22 <4
Exactech Equinoxe[HH.Stand:H.Mod] 125 <4 122 9 <4 8
DePuy Global AP[HH.Stand:H.Mod] 252 5 247 <4 0 <4
Global Advantage[HH.Stand:H.
DePuy 321 66 255 <4 <4 0
Standard]
DePuy Global Unite[HH.Stand:H.Mod] 29 17 12 0 0 0
Global Unite[HH.Stand:H.
DePuy 319 239 80 24 22 <4
NeckBody:H.Mod]
Smith & Nephew Neer[H.MBStem] 24 8 16 0 0 0
Zimmer Biomet Nottingham[HH.Stand:H.Standard] 38 18 20 0 0 0
Corin Oxford[HH.Stand:H.Standard] 76 <4 73 0 0 0
Lima SMR[HH.Stand:H.Dia] 13 8 5 0 0 0
Lima SMR[HH.Stand:H.NeckBody:H.Dia] 298 166 132 17 13 4
JRI Vaios[HH.Stand:H.NeckBody:H.Dia] 86 43 43 5 5 0
Table 3.S5 (d) Number of resurfacing total shoulder replacement replacements between 2012 and 2020 and
within the last year by brand construct.
Acute Acute

Aequalis[G.Ana]: Aequalis
Wright 25 0 25 0 0 0
Resurfacing[HH.Resurf]
Resurfacing TSR
Aequalis Perform+[G.Ana]: Aequalis

Wright 14 0 14 4 0 4
Resurfacing[HH.Resurf]
FH Arrow[G.Ana:HH.Resurf] 15 0 15 0 0 0
DePuy Epoca[G.Ana:HH.Resurf] 126 0 126 0 0 0
DePuy Epoca[G.Peg:G.Ana:HH.Resurf] 54 0 54 0 0 0
DePuy Epoca[G.BP:G.Ana:HH.Resurf] 204 0 204 0 0 0
Exactech Equinoxe[G.Ana:HH.Resurf:H.RPeg] 31 0 31 <4 0 <4
Table 3.S5 (e) Number of stemless conventional total shoulder replacement replacements between 2012 and
2020 and within the last year by brand construct.
All Acute All Acute

Epoca[G.BP]: Epoca[G.Ana]:
DePuy:Mathys 39 0 39 0 0 0
Affinis[HH.Stand]: Affinis[H.Stemless]
Epoca[G.Ana]: Eclipse[HH.Stand]:
Arthrex:DePuy 16 0 16 0 0 0
Eclipse[H.Stemless]
Aequalis[G.Ana]: Eclipse[HH.Stand]:
Arthrex:Wright 77 0 77 0 0 0
Eclipse[H.Stemless]
Universal[G.BP]: Universal[G.Lin]:
Arthrex 63 0 63 12 0 12
Eclipse[HH.Stand]: Eclipse[H.Stemless]
Global Anchor Peg[G.Ana]: Eclipse[HH.
Stand]: Eclipse[H.Stemless]
Epoca[G.Peg]: Epoca[G.Ana]:
Eclipse[HH.Stand]: Eclipse[H.Stemless]
Epoca[G.BP]: Epoca[G.Ana]:
Arthrex:DePuy Eclipse[HH.Stand]: Eclipse[H. 51 0 51 0 0 0
Stemless]
Univers II[G.Ana]: Eclipse[HH.Stand]:
Arthrex 416 0 416 39 0 39
Eclipse[H.Stemless]
Global[G.Ana]: Global ICON[HH.
DePuy 13 0 13 0 0 0
Stand]: Global ICON[H.Stemless]

Global Anchor Peg[G.Ana]: Global
DePuy ICON[HH.Stand]: Global ICON[H. 229 0 229 51 0 51
Stemless]
Comprehensive[G.Peg]:
Stemless TSR
Zimmer Biomet Comprehensive[G.Ana]: Versa- 554 <4 553 56 0 56

Dial[HH.Stand]: Nano[H.Stemless]
TM[G.Ana]: Bigliani/Flatow[HH.
Stand]: Sidus[H.Stemless]
TM[G.Ana]: Sidus[HH.Stand]:
Sidus[H.Stemless]
Comprehensive[G.Peg]:
Zimmer Biomet Comprehensive[G.Ana]: Sidus[HH. 135 0 135 31 0 31
Bigliani/Flatow[G.Ana]: Bigliani/
Flatow[HH.Stand]: Sidus[H.Stemless]
Anatomical[G.Ana]: Sidus[HH.Stand]:
Sidus[H.Stemless]
Bigliani/Flatow[G.Ana]: Sidus[HH.
Aequalis[G.Ana]: Simpliciti[HH.Stand]:
Wright 85 0 85 <4 0 <4
Simpliciti[H.Stemless]
Aequalis Perform+[G.Ana]:
Wright Simpliciti[HH.Stand]: Simpliciti[H. 654 <4 653 69 0 69
Stemless]
Affiniti[G.Ana]: Simpliciti[HH.Stand]:
Wright 10 0 10 0 0 0
Simpliciti[H.Stemless]
Mathys Affinis[G.Ana:HH.Stand:H.Stemless] 1,997 0 1,997 166 0 166
SMR[G.BP:G.Lin:HH.Stand:H.
Lima 150 0 150 17 0 17
Stemless]
Lima SMR[G.Ana:HH.Stand:H.Stemless] 46 0 46 20 0 20
Zimmer Biomet TESS[G.Ana:HH.Stand:H.Stemless] 69 0 69 0 0 0
Table 3.S5 (f) Number of stemmed conventional total shoulder replacements between 2012 and 2020 and within
the last year by brand construct.
Primary operations all Primary operations in
years 2020
All Acute All Acute

Aequalis Perform+[G.Ana]: Aequalis[HH.Stand]:
Wright 50 0 50 0 0 0
Aequalis[H.Standard]
Aequalis[G.Ana]: Aequalis[HH.Stand]: Aequalis-
Wright 10 0 10 0 0 0
Press-Fit[H.Standard]
Aequalis Perform+[G.Ana]: Affiniti[HH.Stand]:
Wright 12 0 12 0 0 0
Affiniti[H.Standard]
TM[G.Ana]: Bigliani/Flatow[HH.Stand]:
Zimmer Biomet 116 0 116 <4 0 <4
Anatomical[H.Mod]
TM Reverse[G.BP]: TM[G.Ana]: Bigliani/
Flatow[HH.Stand]: Anatomical[H.Mod]
Anatomical[G.Ana]: Bigliani/Flatow[HH.Stand]:
Anatomical[H.Mod]
Bigliani/Flatow[G.Ana]: Bigliani/Flatow[HH.
Stand]: Anatomical[H.Mod]
TM[G.Ana]: Anatomical[HH.Stand]:
Anatomical[H.Mod]
Aequalis[G.Ana]: Ascend[HH.Stand]: Ascend[H.
Wright 24 0 24 0 0 0
Standard]
Aequalis[G.Ana]: Ascend Flex[HH.Stand]:
Wright 19 0 19 0 0 0
Ascend Flex[H.Standard]
Aequalis Perform+[G.Ana]: Ascend Flex[HH.
Wright 1,317 0 1,317 143 0 143
Stand]: Ascend Flex[H.Standard]
Comprehensive[G.Ana]: Versa-Dial[HH.Stand]:
Zimmer Biomet 14 0 14 <4 0 <4
Stemmed TSR
Comprehensive[G.Peg]: Comprehensive[G.
Zimmer Biomet Ana]: Versa-Dial[HH.Stand]: Comprehensive[H. 892 <4 890 73 0 73
Standard]
Global[G.Ana]: Global AP[HH.Stand]: Global
DePuy 59 0 59 0 0 0
AP[H.Mod]
Global Anchor Peg[G.Ana]: Global Unite[HH.
DePuy 139 0 139 35 0 35
Stand]: Global AP[H.Mod]
Global Anchor Peg[G.Ana]: Global AP[HH.
DePuy 1,051 0 1,051 9 0 9
Stand]: Global AP[H.Mod]
Global Anchor Peg[G.Ana]: Global
DePuy Advantage[HH.Stand]: Global Advantage[H. 241 0 241 13 0 13
Standard]
Global[G.Ana]: Global Advantage[HH.Stand]:
DePuy 535 0 535 17 0 17
Global Advantage[H.Standard]
Univers II[G.Ana]: Global Unite[HH.Stand]:
Arthrex:DePuy 19 0 19 <4 0 <4
Global Unite[H.NeckBody]: Global Unite[H.Mod]
Global[G.Ana]: Global Unite[HH.Stand]: Global
DePuy 37 0 37 0 0 0
Unite[H.NeckBody]: Global Unite[H.Mod]
DePuy 26 0 26 <4 0 <4
Stand]: Global Unite[H.Mod]
DePuy Stand]: Global Unite[H.NeckBody]: Global 497 <4 496 21 0 21
Unite[H.Mod]
Axioma[G.Peg]: Axioma[G.BP]: SMR[G.Lin]:
Lima SMR[HH.Stand]: SMR[H.NeckBody]: SMR[H. 32 0 32 0 0 0
Dia]
TM[G.Ana]: Bigliani/Flatow[HH.Stand]: TM[H.
Dia]
Table 3.S5 (f) (continued)
Primary operations all Primary operations in

years 2020
All Acute All Acute

Bigliani/Flatow[G.Ana]: Bigliani/Flatow[HH.
Stand]: TM[H.Dia]
Wright Aequalis[G.Ana:HH.Stand:H.Standard] 193 0 193 0 0 0

Mathys Affinis[G.Ana:HH.Stand:H.Standard] 102 <4 101 <4 0 <4
Zimmer Biomet Anatomical[G.Ana:HH.Stand:H.Mod] 85 0 85 0 0 0
FH Arrow[G.Ana:HH.Stand:H.Standard] 165 0 165 <4 0 <4
FH Arrow[G.BP:G.Lin:HH.Stand:H.Standard] 14 0 14 <4 0 <4
Stemmed TSR
Zimmer Biomet Bigliani/Flatow[G.Ana:HH.Stand:H.Dia] 58 0 58 0 0 0

DePuy Epoca[G.BP:G.Ana:HH.Stand:H.Mod] 62 <4 60 0 0 0
DePuy Epoca[G.Ana:HH.Stand:H.Mod] 315 0 315 0 0 0
DePuy Epoca[G.Peg:G.Ana:HH.Stand:H.Mod] 156 0 156 0 0 0
Exactech Equinoxe[G.Ana:HH.Stand:H.Mod] 1,158 <4 1,156 89 0 89
Medacta[G.Ana:HH.Stand:H.NeckBody:H.
Medacta 18 0 18 <4 0 <4
Standard]
Lima SMR[G.BP:G.Lin:HH.Stand:H.NeckBody:H.Dia] 407 0 407 6 0 6
Lima SMR[G.Ana:HH.Stand:H.NeckBody:H.Dia] 50 0 50 5 0 5
Vaios[G.BP:G.Ana:HH.Stand:H.NeckBody:H.
JRI 125 0 125 <4 0 <4
Dia]
Table 3.S5 (g) Number of stemless reverse polarity total shoulder replacements between 2012 and 2020 and
within the last year by brand construct.

All Acute All Acute
Comprehensive[G.BP]: Versa-Dial[G.
Stemless RTSR
Zimmer Biomet Sph]: Comprehensive[H.RevBear]: 37 0 37 0 0 0

Nano[H.Stemless]
SMR[G.BP:G.Sph:H.RevBear:H.
Lima 137 0 137 20 0 20
Stemless]
TESS[G.BP:G.Sph:H.RevBear:H.
Stemless]
Table 3.S5 (h) Number of stemmed reverse polarity total shoulder replacement replacements between 2012
and 2020 and within the last year by brand construct.
All Acute All Acute

Aequalis Perform Reversed[G.BP]: Aequalis
Perform Reversed[G.Sph]: Aequalis Reversed
Wright 37 25 12 15 9 6
Fracture[H.RevBear]: Aequalis Reversed
Aequalis-Reversed II[G.BP]: Aequalis-Reversed
Wright II[G.Sph]: Aequalis Reversed Fracture[H.RevBear]: 57 39 18 7 5 <4
Aequalis Reversed Fracture[H.Standard]
Wright II[G.Sph]: Aequalis-Reversed II[H.RevBear]: 372 285 87 44 39 5
Aequalis Reversed Fracture[H.Standard]
Wright Perform Reversed[G.Sph]: Aequalis-Reversed II[H. 85 67 18 38 36 <4
RevBear]: Aequalis Reversed Fracture[H.Standard]
II[G.Sph]: Aequalis-Reversed II[H.RevBear]:
Wright 10 8 <4 <4 <4 0
Aequalis Reversed Fracture[H.Spacer]: Aequalis
Reversed Fracture[H.Standard]
Perform Reversed[G.Sph]: Aequalis-Reversed

Wright 122 8 114 28 <4 24
II[H.RevBear]: Aequalis-Reversed II[H.RevCup]:
Aequalis-Reversed II[H.Dia]
TM Reverse[G.BP]: TM Reverse[G.Sph]:
Zimmer Biomet 1,082 37 1,045 63 <4 61
Anatomical I/R[H.RevBear]: Anatomical[H.Mod]
Anatomical I/R[G.BP]: Anatomical I/R[G.Sph]:
Anatomical I/R[H.RevBear]: Anatomical[H.Mod]
Stemmed RTSR
TM Reverse[G.BP]: TM Reverse[G.Sph]:
Zimmer Biomet Anatomical I/R[H.RevBear]: Anatomical 131 107 24 17 17 0
Fracture[H.Mod]
Wright II[G.Sph]: Ascend Flex[H.RevBear]: Ascend Flex[H. 12 <4 11 0 0 0
Standard]
Wright II[G.Sph]: Ascend Flex[H.RevBear]: Ascend Flex[H. 1,441 17 1,424 157 4 153
RevCup]: Ascend Flex[H.Standard]
Perform Reversed[G.Sph]: Ascend Flex[H.
Wright 1,051 35 1,016 273 16 257
RevBear]: Ascend Flex[H.RevCup]: Ascend Flex[H.
Standard]
Aequalis Perform Reversed[G.BP]: Unbranded[G.
Wright Sph]: Ascend Flex[H.RevBear]: Ascend Flex[H. 21 0 21 8 0 8
RevCup]: Ascend Flex[H.Standard]
Comprehensive[G.BP]: Versa-Dial Glenosphere[G.
Zimmer Biomet Sph]: Comprehensive[H.RevBear]: 12 <4 11 <4 0 <4
Comprehensive[G.BP]: Versa-Dial[G.Sph]:
Zimmer Biomet Comprehensive[H.RevBear]: Comprehensive[H. 2,408 90 2,318 213 12 201
Standard]
Zimmer Biomet Comprehensive[H.RevBear]: Comprehensive 475 378 97 66 53 13
Comprehensive[H.RevBear]: Comprehensive
Zimmer Biomet 19 5 14 <4 <4 0
Segmental Revision[H.NeckBody]: Comprehensive
Segmental Revision[H.Dia]
Note: HH.=Humeral head, H.=Humerus, G.=Glenoid, Resurf= Resurfacing, RPeg=Resurfacing peg, Ana=Anatomic, BP=Baseplate, Peg=Peg, Stand=Standard,
Lin=Liner, Sph=Sphere, RevBear=Reverse bearing, Stand=Standard, NeckBody=Modular neck body, Mod=Modular Stem, MBStem=Monobloc stem,
Dia=Diaphyseal stem, RevBear=Reverse bearing, RevCup=Reverse cup.
Note: Data is sorted by the brand of the humeral component.
Table 3.S5 (h) (continued)
All Acute All Acute

Delta Xtend[G.BP]: Delta Xtend[G.Sph]: Delta
DePuy Xtend[H.RevBear]: Delta Xtend[H.RevCup]: Global 94 59 35 18 11 7
Unite[H.Mod]
Axioma[G.Peg]: Axioma[G.BP]: SMR[G.Sph]:
Lima 101 <4 99 10 0 10
SMR[H.RevBear]: SMR[H.RevCup]: SMR[H.Dia]
Axioma[G.BP]: SMR[G.Sph]: SMR[H.RevBear]:
Lima 94 <4 91 0 0 0
SMR[H.RevCup]: SMR[H.Dia]
Comprehensive[G.BP]: Versa-Dial[G.Sph]: TM
Zimmer Biomet 39 0 39 <4 0 <4
Reverse[H.RevBear]: TM Reverse[H.Mod]
Universal[G.BP]: Univers Reverse[G.Sph]: Univers
Arthrex Reverse[H.RevBear]: Univers Reverse[H.RevCup]: 47 7 40 7 <4 5
Univers Reverse[H.Standard]
Arthrex Reverse[H.RevBear]: Univers Reverse[H.Spacer]: 11 <4 10 0 0 0
Univers Reverse[H.Standard]
Arthrex 184 18 166 14 <4 12
Reverse[H.RevBear]: Univers Reverse[H.Standard]
Aequalis-Reversed II[G.BP:G.Sph:H.RevBear:H.
Wright 16 0 16 <4 0 <4
RevCup:H.Spacer:H.Dia]

Wright 19 0 19 <4 0 <4
Dia]
Wright 1,173 25 1,148 46 <4 43
RevCup:H.Dia]
Mathys Affinis[G.BP:G.Sph:H.RevBear:H.Standard] 817 31 786 57 <4 56
Stemmed RTSR
Mathys Affinis[G.BP:G.Sph:H.RevBear:H.Dia] 173 134 39 20 17 <4

Affinis[G.BP:G.Sph:H.RevBear:H.Spacer:H.
Mathys 15 <4 13 0 0 0
Standard]
FH Arrow[G.BP:G.Sph:H.RevBear:H.Standard] 169 24 145 9 <4 8
Delta Xtend[G.BP:G.Sph:H.RevBear:H.RevCup:H.
DePuy 22 <4 19 0 0 0
Spacer:H.Mod]
Delta Xtend[G.BP:G.Sph:H.RevBear:H.RevCup:H.
DePuy 2,620 64 2,556 189 14 175
Mod]
DePuy Delta Xtend[G.BP:G.Sph:H.RevBear:H.Standard] 2,972 538 2,434 180 54 126
Delta Xtend[G.BP:G.Sph:H.RevBear:H.Spacer:H.
DePuy 84 32 52 <4 0 <4
Standard]
DePuy Delta Xtend[G.BP:G.Sph:H.RevBear:H.Mod] 43 4 39 <4 <4 <4
Exactech Equinoxe[G.BP:G.Sph:H.RevBear:H.Standard] 415 329 86 65 52 13
Exactech Equinoxe[G.BP:G.Sph:H.RevBear:H.Mod] 2,772 50 2,722 287 11 276
Stanmore METS[G.Sph:H.RevBear:H.Mod] 11 0 11 0 0 0
DJO RSP[G.BP:G.Sph:H.RevBear:H.Mod] 28 <4 25 0 0 0
DJO RSP[G.BP:G.Sph:H.RevBear:H.Standard] 451 38 413 79 6 73
RSP[G.BP:G.Sph:H.RevBear:H.Spacer:H.
DJO 9 <4 8 5 <4 4
Standard]
Lima SMR[G.BP:G.Sph:H.RevBear:H.RevCup:H.Dia] 1,806 332 1,474 198 57 141
SMR[G.BP:G.Sph:H.RevBear:H.RevCup:H.
Lima 154 35 119 9 <4 6
Spacer:H.Dia]
TM Reverse[G.BP:G.Sph:H.RevBear:H.Spacer:H.
Mod]
Zimmer Biomet TM Reverse[G.BP:G.Sph:H.RevBear:H.Mod] 656 65 591 48 7 41
JRI Vaios[G.BP:G.Sph:H.RevBear:H.NeckBody:H.Dia] 357 28 329 11 <4 9
Innovative Verso[G.BP:G.Sph:H.RevBear:H.Standard] 615 39 576 69 <4 68
Note: HH.=Humeral head, H.=Humerus, G.=Glenoid, Resurf= Resurfacing, RPeg=Resurfacing peg, Ana=Anatomic, BP=Baseplate, Peg=Peg, Stand=Standard,
Lin=Liner, Sph=Sphere, RevBear=Reverse bearing, Stand=Standard, NeckBody=Modular neck body, Mod=Modular Stem, MBStem=Monobloc stem,
Dia=Diaphyseal stem, RevBear=Reverse bearing, RevCup=Reverse cup.
Tables 3.S5 (a) to (h) on the previous pages illustrate facilitate improved implant scrutiny. Given the rapid
the shoulder construct used by sub-type of the evolution and heterogeneity of shoulder prostheses,
primary shoulder replacement for overall procedures it is expected that the classification system will evolve
and by acute and elective sub-divisions. They also year on year with the introduction of new types of
show this data for the last year. Implants are only prostheses and the combinations in which these are
listed if they have been used on more than ten used by surgeons.
occasions overall or five occasions within the last
year respectively. Results illustrate the frequency 3.6.2 Revisions after primary shoulder
of all implanted constructs across all years of data replacement surgery
collection within the registry i.e. between 2012 and
2020. The frequency of shoulder constructs within We present results in this section as percentage
the last year of the data collection is also illustrated cumulative revision of primary shoulder replacements.
to indicate contemporary practice. Constructs and Results are estimated using the 1-Kaplan-Meier
prostheses elements are suffixed ‘[ ]’ to indicate the method; 95% CIs are shown within tables and when
implants that make up the construct. In the cases of number at risk falls below 250, estimates are shown
within manufacturer and brand construct, this suffix in blue italics to indicate that caution is required in
is placed after the brand name; whereas within mix interpreting the results. Data are presented up to eight
and match constructs, the suffix is placed immediately years which is the last full year of data collection within
after the brand of the implanted element. While the the registry. Figures also include an ‘at-risk table’
detail in reporting of constructs has become more which presents the number of individuals at risk of
granular, the complexity has necessarily increased revision at the time indicated.
to reflect the diversity of implanted elements and will
Figure 3.S5 KM estimates of cumulative revision for primary shoulder replacement by acute trauma and
elective cases. Blue italics in the numbers at risk table signify that fewer than 250 cases remained at risk at
these time points.
10

0
0 1 2 3 4 5 6 7 8
Years since primary
Acute trauma 5,143 4,256 3,269 2,437 1,715 1,129 682 350 86
Elective 45,112 40,786 33,169 26,120 19,553 13,711 8,908 4,750 1,641
Table 3.S6 KM estimates of cumulative revision (95% CI) for primary shoulder replacement for all cases, acute
trauma and elective cases. Blue italics signify that fewer than 250 cases remained at risk at these time points.
Age at
primary Percentage Time since primary
Median male
(IQR) (%) 1 year 2 years 3 years 4 years 5 years 6 years 7 years 8 years
All 73 1.37 2.39 3.15 3.80 4.32 4.78 5.33 5.68
50,255 30
cases (67 to 79) (1.27-1.48) (2.25-2.53) (2.99-3.33) (3.61-4.00) (4.11-4.55) (4.54-5.03) (5.04-5.64) (5.35-6.04)
Acute 74 1.43 2.51 2.88 3.31 3.65 3.97 3.97 4.27
5,143 23
trauma (67 to 80) (1.14-1.81) (2.09-3.02) (2.41-3.43) (2.78-3.95) (3.05-4.36) (3.29-4.78) (3.29-4.78) (3.42-5.32)
73 1.36 2.37 3.18 3.84 4.38 4.85 5.44 5.80
Elective 45,112 31
(67 to 79) (1.26-1.48) (2.23-2.53) (3.00-3.36) (3.64-4.05) (4.16-4.62) (4.60-5.12) (5.14-5.77) (5.44-6.17)
Figure 3.S5 and Table 3.S6 illustrate the cumulative years following surgery, at which point it starts to
revision of primary shoulder procedures performed diverge. The risk of revision for acute trauma patients
overall (shown in Table 3.S6 only) and by acute tends to be lower, but the number of patients still at
trauma and elective procedures. Our results indicate risk at eight years is small and therefore should be
that the risk of revision is comparable for the first two interpreted cautiously.
Table 3.S7 KM estimates of cumulative revision (95% CI) for primary shoulder replacement for elective cases by
gender and age group. Blue italics signify that fewer than 250 cases remained at risk at these time points.
primary
Gender (years) N 1 year 2 years 3 years 4 years 5 years 6 years 7 years 8 years
0.98 1.88 2.60 3.15 3.65 4.08 4.57 5.00
All 31,351
(0.87-1.10) (1.73-2.05) (2.42-2.80) (2.93-3.38) (3.40-3.91) (3.80-4.38) (4.23-4.93) (4.59-5.45)
2.42 4.82 7.50 8.78 10.07 10.40 11.91 13.27
<55 1,146
(1.67-3.51) (3.68-6.29) (6.00-9.35) (7.11-10.80) (8.21-12.32) (8.46-12.76) (9.47-14.92) (10.35-16.93)
Female
1.37 2.80 4.04 5.10 6.40 7.63 8.72 9.20

55 to 64 3,263
(1.02-1.84) (2.26-3.47) (3.36-4.86) (4.29-6.05) (5.42-7.54) (6.47-8.98) (7.38-10.29) (7.74-10.93)
1.01 2.02 2.84 3.41 3.89 4.38 4.96 5.54
65 to 74 11,559
(0.84-1.21) (1.77-2.31) (2.53-3.19) (3.05-3.81) (3.49-4.34) (3.93-4.90) (4.41-5.58) (4.84-6.34)
0.76 1.35 1.72 2.05 2.30 2.47 2.59 2.77
≥75 15,383
(0.64-0.92) (1.18-1.56) (1.51-1.96) (1.81-2.33) (2.03-2.61) (2.17-2.81) (2.27-2.97) (2.37-3.22)
2.24 3.50 4.50 5.46 6.09 6.65 7.49 7.64
All 13,761
(2.00-2.50) (3.19-3.84) (4.13-4.89) (5.03-5.91) (5.62-6.59) (6.13-7.22) (6.85-8.20) (6.97-8.37)
2.63 5.39 7.36 9.73 11.25 12.60 14.83 15.31
<55 1,479
(1.92-3.60) (4.31-6.74) (6.04-8.95) (8.12-11.63) (9.45-13.38) (10.56-15.00) (12.27-17.87) (12.63-18.51)
2.03 3.49 4.84 5.84 6.39 7.47 8.19 8.19
Male
55 to 64 2,614
(1.55-2.66) (2.82-4.32) (4.01-5.84) (4.88-6.97) (5.36-7.62) (6.24-8.93) (6.76-9.90) (6.76-9.90)
2.09 3.09 3.78 4.68 5.48 5.81 6.50 6.50
65 to 74 5,241
(1.73-2.52) (2.64-3.62) (3.27-4.38) (4.07-5.38) (4.77-6.28) (5.06-6.68) (5.59-7.55) (5.59-7.55)
2.41 3.33 4.12 4.56 4.63 4.75 5.13 5.46
≥75 4,427
(1.99-2.91) (2.82-3.93) (3.53-4.82) (3.91-5.32) (3.97-5.40) (4.06-5.55) (4.29-6.13) (4.45-6.68)
Table 3.S7 further breaks down the cumulative revision

of primary shoulder procedures for elective patients, by
gender and age group. Results indicate that females
have a lower risk of revision in the long term compared
to males and that younger patients have an increased
risk of revision compared to older patients.
Figure 3.S6 KM estimates of cumulative revision for primary elective shoulder replacement by type of
shoulder replacement. Blue italics in the numbers at risk table signify that fewer than 250 cases remained
15
12

6
0
0 1 2 3 4 5 6 7 8
Years since primary
Resurfacing HHA 2,900 2,802 2,571 2,321 2,019 1,599 1,229 773 322
Stemless HHA 1,217 1,155 955 753 575 408 282 149 46
Stemmed HHA 2,299 2,155 1,852 1,572 1,296 1,004 694 417 162
Resurfacing TSR 486 474 449 412 358 280 200 124 37
Stemless TSR 4,911 4,378 3,391 2,521 1,767 1,156 680 328 110
Stemmed TSR 8,323 7,745 6,635 5,507 4,246 3,076 1,989 1,027 346
Stemless RTSR 186 154 130 91 71 47 25 11 <4
Stemmed RTSR 20,481 18,004 13,810 10,196 7,037 4,527 2,709 1,279 386
IPA 5 5 <4
Note: HHA=Humeral hemiarthroplasty, TSR=Total shoulder replacement, RTSR=Reverse polarity total shoulder replacement, IPA=Interpositional arthroplasty.
280
Table 3.S8 KM estimates of cumulative revision (95% CI) for primary shoulder replacement for elective cases by shoulder type.
Age at
primary Percentage Time since primary
Median male
Elective N (IQR) (%) 1 year 2 years 3 years 4 years 5 years 6 years 7 years 8 years
Proximal humeral 70 0.87 3.10 4.95 6.32 7.65 8.48 9.47 9.91
6,416 33
hemiarthroplasty (61 to 77) (0.67-1.14) (2.69-3.57) (4.41-5.55) (5.70-7.00) (6.94-8.42) (7.71-9.32) (8.60-10.42) (8.99-10.93)
71 0.66 3.04 4.61 6.25 7.93 8.96 10.04 10.53
Resurfacing 2,900 31
(64 to 78) (0.42-1.04) (2.46-3.75) (3.88-5.48) (5.38-7.26) (6.92-9.09) (7.85-10.21) (8.81-11.43) (9.23-12.00)
67 0.76 3.04 5.01 6.51 8.32 9.81 11.95 11.95
Stemless 1,217 42
(56 to 75) (0.39-1.45) (2.18-4.23) (3.84-6.54) (5.10-8.29) (6.61-10.46) (7.81-12.29) (9.38-15.16) (9.38-15.16)
70 1.21 3.18 5.36 6.27 6.83 7.05 7.43 7.95
Stemmed 2,299 32
(60 to 78) (0.83-1.75) (2.51-4.03) (4.45-6.45) (5.26-7.47) (5.75-8.09) (5.94-8.36) (6.23-8.86) (6.59-9.59)
Total shoulder 70 0.94 1.86 2.56 3.19 3.58 3.93 4.40 4.74
13,720 31
replacement (64 to 76) (0.79-1.12) (1.64-2.11) (2.29-2.86) (2.87-3.55) (3.22-3.97) (3.53-4.37) (3.92-4.94) (4.17-5.39)
71 0.62 1.47 2.16 2.92 3.27 3.73 4.48 4.48
Resurfacing 486 29
(63 to 76) (0.20-1.91) (0.70-3.06) (1.17-3.97) (1.70-5.00) (1.93-5.49) (2.22-6.22) (2.61-7.64) (2.61-7.64)
69 0.73 1.60 2.17 2.81 3.01 3.38 3.76 3.76
Stemless 4,911 36
(62 to 75) (0.52-1.02) (1.26-2.02) (1.75-2.68) (2.30-3.44) (2.45-3.68) (2.71-4.20) (2.95-4.79) (2.95-4.79)
71 1.09 2.04 2.80 3.41 3.88 4.21 4.68 5.17
Stemmed 8,323 29
(65 to 76) (0.88-1.34) (1.74-2.38) (2.44-3.20) (3.00-3.88) (3.42-4.39) (3.71-4.77) (4.08-5.35) (4.43-6.02)
Reverse polarity
76 1.67 2.28 2.66 2.95 3.13 3.34 3.61 3.85
total shoulder 20,667 29
(71 to 80) (1.50-1.86) (2.07-2.50) (2.43-2.90) (2.70-3.23) (2.86-3.42) (3.04-3.67) (3.23-4.02) (3.36-4.41)
replacement
73 5.52 6.18 6.96 6.96 8.31 8.31 8.31
Stemless 186 38
(69 to 78) (3.00-10.01) (3.46-10.89) (3.99-11.99) (3.99-11.99) (4.75-14.33) (4.75-14.33) (4.75-14.33)
76 1.64 2.24 2.62 2.92 3.08 3.29 3.56 3.81

Stemmed 20,481 28
(71 to 80) (1.47-1.82) (2.04-2.46) (2.39-2.86) (2.66-3.19) (2.81-3.37) (2.99-3.62) (3.19-3.98) (3.32-4.37)
Interpositional 58
5 60
arthroplasty (55 to 68)
73 1.96 3.12 4.16 4.97 5.57 6.30 7.07 7.45
Unconfirmed 4,304 33
(66 to 78) (1.58-2.42) (2.62-3.71) (3.56-4.85) (4.29-5.74) (4.82-6.42) (5.46-7.27) (6.10-8.19) (6.37-8.71)
69 1.03 3.86 5.29 5.84 5.84 6.65 8.06 8.06
Unconfirmed HHA 292 37
(58.5 to 75.5) (0.33-3.17) (2.09-7.07) (3.09-8.97) (3.48-9.73) (3.48-9.73) (3.99-10.97) (4.70-13.65) (4.70-13.65)
69 1.12 2.72 4.19 5.29 6.30 7.19 8.01 8.40
Unconfirmed TSR 1,814 36
(61 to 76) (0.73-1.74) (2.05-3.61) (3.33-5.27) (4.29-6.51) (5.17-7.67) (5.92-8.72) (6.59-9.73) (6.83-10.30)
75 2.78 3.27 3.80 4.29 4.43 4.92 5.44 5.91
Unconfirmed RTSR 2,195 29
(69 to 80) (2.16-3.58) (2.59-4.14) (3.03-4.76) (3.43-5.36) (3.54-5.54) (3.91-6.18) (4.25-6.95) (4.49-7.76)
60
Unconfirmed IPA <4 100
(58 to 65)
Table 3.S8 and Figure 3.S6 report cumulative The cumulative risk of revision of stemless reverse
revision of primary shoulder procedures, for elective polarity total shoulder replacements is higher
patients, by type (Table 3.S8 only) and sub-type of compared to stemmed versions. This needs
shoulder construct. careful interpretation as the number of stemless
reverse polarity replacements is low, however, it is
Proximal humeral hemiarthroplasties undergo worth noting that some stemless reverse polarity
revision at a higher rate than either conventional brands have been withdrawn from the market. The
total shoulder replacements or reverse polarity total performance of stemmed conventional total shoulder
shoulder replacements. The extent to which proximal replacement compared to stemmed reverse polarity
humeral hemiarthroplasty procedures are seen as shoulder replacements is of particular interest. Reverse
‘revisable’ procedures compared to total shoulder polarity total shoulder replacements tend to have
replacements should be considered when interpreting an initially higher revision rate which then plateaus,
the results. Furthermore, while Table 3.S8 and whereas the conventional total shoulder replacements
Figure 3.S6 suggest a stemmed proximal humeral increase more slowly but at a constant rate and
hemiarthroplasty might be the better choice over a therefore exceed the cumulative risk of revision of
stemless or resurfacing humeral hemiarthroplasty, the reverse polarity total replacements and overall is 0.9%
latter group are more straightforward to revise than higher at eight years. The extent to which the different
a stemmed implant and so caution is again needed indications for surgery are confounding results is not
interpreting these sub-group results. clear and results should be interpreted cautiously.
282
Table 3.S9 KM estimates of cumulative revision (95% CI) for primary shoulder replacement for elective cases by brand construct in constructs
with greater than 250 implantations. Blue italics signify that fewer than 250 cases remained at risk at these time points.
Time since primary
Shoulder construct N 1 year 2 years 3 years 4 years 5 years 6 years 7 years 8 years
0.40 2.43 3.69 4.59 6.17 6.82 8.60 8.60
Aequalis Resurfacing[HH.Resurf] 251
(0.06-2.81) (1.10-5.33) (1.94-6.97) (2.57-8.14) (3.68-10.24) (4.14-11.14) (5.30-13.79) (5.30-13.79)
Resurfacing 0.44 2.26 3.66 5.10 6.89 7.76 9.12 9.89
Copeland[HH.Resurf] 1,617
HHA (0.21-0.92) (1.63-3.13) (2.82-4.74) (4.08-6.37) (5.65-8.39) (6.41-9.37) (7.58-10.95) (8.22-11.89)
1.00 3.67 5.21 7.30 8.74 10.34 10.73 10.73
Global CAP[HH.Resurf] 607
(0.45-2.21) (2.41-5.57) (3.65-7.41) (5.38-9.87) (6.58-11.57) (7.88-13.52) (8.18-14.01) (8.18-14.01)
Stemless 0.35 2.84 3.07 4.66 6.55 8.04 9.32 9.32
Affinis[HH.Stand:H.Stemless] 581
HHA (0.09-1.41) (1.72-4.66) (1.89-4.96) (3.02-7.17) (4.33-9.85) (5.28-12.15) (5.95-14.44) (5.95-14.44)
Stemmed 1.18 1.60 4.21 4.69 5.20 5.76 5.76 5.76
Global Advantage[HH.Stand:H.Standard] 255
HHA (0.38-3.62) (0.60-4.20) (2.28-7.68) (2.62-8.31) (2.98-8.99) (3.37-9.75) (3.37-9.75) (3.37-9.75)
Univers II[G.Ana]: Eclipse[HH.Stand]: Eclipse[H. 0.24 1.40 2.49 3.00 3.00 3.00 3.00 3.00
416
Stemless] (0.03-1.69) (0.58-3.34) (1.24-4.95) (1.54-5.79) (1.54-5.79) (1.54-5.79) (1.54-5.79) (1.54-5.79)
Comprehensive[G.Peg]: Comprehensive[G.Ana]: 1.13 2.05 2.92 3.74 4.99 4.99 4.99
553
Stemless Versa-Dial[HH.Stand]: Nano[H.Stemless] (0.51-2.49) (1.11-3.80) (1.69-5.02) (2.22-6.27) (2.96-8.36) (2.96-8.36) (2.96-8.36)
TSR Aequalis Perform+[G.Ana]: Simpliciti[HH.Stand]: 0.50 1.69 1.95 1.95 1.95 1.95
653
Simpliciti[H.Stemless] (0.16-1.54) (0.88-3.23) (1.04-3.61) (1.04-3.61) (1.04-3.61) (1.04-3.61)
0.37 0.79 1.17 1.39 1.54 1.82 2.28 2.28
Affinis[G.Ana:HH.Stand:H.Stemless] 1,997
(0.17-0.77) (0.47-1.34) (0.74-1.84) (0.89-2.16) (0.99-2.38) (1.13-2.94) (1.32-3.93) (1.32-3.93)
Aequalis Perform+[G.Ana]: Ascend Flex[HH. 0.24 0.72 1.39 2.27 2.27 2.27
1,317
Stand]: Ascend Flex[H.Standard] (0.08-0.76) (0.36-1.44) (0.80-2.41) (1.36-3.78) (1.36-3.78) (1.36-3.78)
Comprehensive[G.Peg]: Comprehensive[G.
1.52 2.85 4.42 4.88 4.88 5.32 5.32 5.32
Ana]: Versa-Dial[HH.Stand]: Comprehensive[H. 890

(0.89-2.61) (1.90-4.27) (3.13-6.23) (3.48-6.83) (3.48-6.83) (3.75-7.52) (3.75-7.52) (3.75-7.52)
Standard]
Global Anchor Peg[G.Ana]: Global AP[HH.Stand]: 0.29 0.78 1.08 1.43 1.59 2.12 2.12 2.12
1,051
Global AP[H.Mod] (0.09-0.89) (0.39-1.55) (0.60-1.95) (0.85-2.41) (0.95-2.63) (1.32-3.39) (1.32-3.39) (1.32-3.39)
Global[G.Ana]: Global Advantage[HH.Stand]: 0.57 0.98 1.43 1.97 2.60 2.60 2.60 2.60
535
Stemmed Global Advantage[H.Standard] (0.18-1.76) (0.41-2.34) (0.68-2.99) (1.02-3.77) (1.43-4.71) (1.43-4.71) (1.43-4.71) (1.43-4.71)
TSR Global Anchor Peg[G.Ana]: Global Unite[HH.
0.82 1.73 1.73 1.73 2.30 2.30
Stand]: Global Unite[H.NeckBody]: Global 496
(0.31-2.17) (0.87-3.43) (0.87-3.43) (0.87-3.43) (1.13-4.66) (1.13-4.66)
Unite[H.Mod]
0.32 0.65 1.30 2.03 2.03 2.03 2.84 2.84
Epoca[G.Ana:HH.Stand:H.Mod] 315
(0.04-2.23) (0.16-2.56) (0.49-3.44) (0.92-4.47) (0.92-4.47) (0.92-4.47) (1.28-6.23) (1.28-6.23)
1.18 2.22 3.10 3.99 4.42 4.42 5.62 5.62
Equinoxe[G.Ana:HH.Stand:H.Mod] 1,156
(0.68-2.02) (1.48-3.33) (2.17-4.42) (2.85-5.58) (3.17-6.15) (3.17-6.15) (3.78-8.31) (3.78-8.31)
2.98 5.60 7.94 9.01 9.95 9.95 9.95 14.56
SMR[G.BP:G.Lin:HH.Stand:H.NeckBody:H.Dia] 407
(1.70-5.19) (3.72-8.38) (5.61-11.18) (6.47-12.48) (7.19-13.68) (7.19-13.68) (7.19-13.68) (8.97-23.17)
Note: HH.=Humeral head, H.=Humerus, G.=Glenoid, Resurf= Resurfacing, RPeg=Resurfacing peg, Ana=Anatomic, BP=Baseplate, Peg=Peg, Stand=Standard, Lin=Liner, Sph=Sphere, RevBear=Reverse
bearing, Stand=Standard, NeckBody=Modular neck body, Mod=Modular Stem, MBStem=Monobloc stem, Dia=Diaphyseal stem, RevBear=Reverse bearing, RevCup=Reverse cup.
Table 3.S9 (continued)
Time since primary
Shoulder construct N 1 year 2 years 3 years 4 years 5 years 6 years 7 years 8 years
TM Reverse[G.BP]: TM Reverse[G.Sph]: 1.85 2.41 3.26 3.61 4.17 4.17 4.17 4.17
1,045
Anatomical I/R[H.RevBear]: Anatomical[H.Mod] (1.19-2.89) (1.62-3.58) (2.28-4.65) (2.54-5.10) (2.92-5.92) (2.92-5.92) (2.92-5.92) (2.92-5.92)
Perform Reversed[G.Sph]: Ascend Flex[H. 1.58 2.00 2.47
1,016
RevBear]: Ascend Flex[H.RevCup]: Ascend (0.95-2.61) (1.21-3.29) (1.43-4.25)
Flex[H.Standard]
1.31 1.91 2.02 2.02 2.02 3.44 5.41
II[G.Sph]: Ascend Flex[H.RevBear]: Ascend 1,424
(0.83-2.08) (1.29-2.82) (1.38-2.96) (1.38-2.96) (1.38-2.96) (1.84-6.39) (2.40-11.97)
Flex[H.RevCup]: Ascend Flex[H.Standard]
1.20 1.46 1.68 1.77 1.89 1.89 1.89 1.89
Comprehensive[H.RevBear]: Comprehensive[H. 2,318
(0.82-1.74) (1.04-2.06) (1.20-2.34) (1.27-2.46) (1.35-2.65) (1.35-2.65) (1.35-2.65) (1.35-2.65)
Standard]
Aequalis-Reversed II[G.BP:G.Sph:H.RevBear:H. 1.25 1.72 1.94 2.09 2.09 2.09 2.69 3.97
1,148
RevCup:H.Dia] (0.74-2.10) (1.10-2.68) (1.27-2.97) (1.37-3.16) (1.37-3.16) (1.37-3.16) (1.56-4.62) (1.91-8.15)
3.14 4.24 4.86 5.48 6.26 6.26 7.64 7.64
Affinis[G.BP:G.Sph:H.RevBear:H.Standard] 786
Stemmed (2.11-4.64) (3.00-5.99) (3.48-6.76) (3.93-7.63) (4.47-8.75) (4.47-8.75) (4.89-11.85) (4.89-11.85)
RTSR Delta Xtend[G.BP:G.Sph:H.RevBear:H. 1.13 1.62 1.73 1.73 1.73 1.87 2.18 2.18
2,556
RevCup:H.Mod] (0.78-1.63) (1.19-2.21) (1.28-2.35) (1.28-2.35) (1.28-2.35) (1.36-2.57) (1.48-3.21) (1.48-3.21)
1.21 1.51 1.51 1.58 1.79 2.30 2.30 2.30
Delta Xtend[G.BP:G.Sph:H.RevBear:H.Standard] 2,434

(0.84-1.74) (1.08-2.10) (1.08-2.10) (1.14-2.19) (1.28-2.49) (1.60-3.28) (1.60-3.28) (1.60-3.28)
1.39 2.10 2.69 3.74 4.22 4.47 4.95 4.95
Equinoxe[G.BP:G.Sph:H.RevBear:H.Mod] 2,722
(1.00-1.92) (1.59-2.77) (2.08-3.49) (2.92-4.79) (3.27-5.44) (3.44-5.82) (3.65-6.68) (3.65-6.68)
1.58 2.25 2.25 2.25 2.25 2.25
RSP[G.BP:G.Sph:H.RevBear:H.Standard] 413
(0.71-3.49) (1.13-4.47) (1.13-4.47) (1.13-4.47) (1.13-4.47) (1.13-4.47)
1.73 2.74 3.19 3.32 3.32 3.32 3.32 3.32
SMR[G.BP:G.Sph:H.RevBear:H.RevCup:H.Dia] 1,474
(1.18-2.56) (1.99-3.77) (2.35-4.33) (2.45-4.50) (2.45-4.50) (2.45-4.50) (2.45-4.50) (2.45-4.50)
1.06 1.46 1.97 2.68 2.68 2.68 2.68
TM Reverse[G.BP:G.Sph:H.RevBear:H.Mod] 591
(0.48-2.34) (0.73-2.90) (1.06-3.67) (1.49-4.78) (1.49-4.78) (1.49-4.78) (1.49-4.78)
Vaios[G.BP:G.Sph:H.RevBear:H.NeckBody:H. 2.77 4.12 4.87 4.87 4.87 4.87 4.87 4.87
329
Dia] (1.45-5.26) (2.41-6.99) (2.96-7.97) (2.96-7.97) (2.96-7.97) (2.96-7.97) (2.96-7.97) (2.96-7.97)
2.34 3.23 3.50 3.97 3.97 3.97 3.97 3.97
Verso[G.BP:G.Sph:H.RevBear:H.Standard] 576
(1.36-3.99) (2.01-5.16) (2.21-5.53) (2.49-6.30) (2.49-6.30) (2.49-6.30) (2.49-6.30) (2.49-6.30)
Note: HH.=Humeral head, H.=Humerus, G.=Glenoid, Resurf= Resurfacing, RPeg=Resurfacing peg, Ana=Anatomic, BP=Baseplate, Peg=Peg, Stand=Standard, Lin=Liner, Sph=Sphere, RevBear=Reverse
bearing, Stand=Standard, NeckBody=Modular neck body, Mod=Modular Stem, MBStem=Monobloc stem, Dia=Diaphyseal stem, RevBear=Reverse bearing, RevCup=Reverse cup.
283
Table 3.S9 reports cumulative revision of primary different product lines, the prosthesis is indicated in
shoulder procedures for elective patients by shoulder [ ] immediately after. The description of constructs
construct. All constructs that have been used on more is necessarily complex, this reflects the extensive
than 250 occasions are reported. Where the construct modularity of modern shoulder prostheses. All results
is solely built from within the same product line the should be viewed in the context of observational
elements used to build the construct are suffixed in [ ] data and due consideration given to the volume of
following the brand. Where the construct is built from unconfirmed prostheses.
Table 3.S10 PTIR estimates of indications for shoulder revision (95% CI) for acute trauma by type of shoulder
replacement between 2012 and 2020.
Number of revisions per 100 prosthesis-years at risk for:
insufficiency
Instablility |
Dislocation
indications
loosening |
All causes
prosthetic
Prosthesis-
Infection
fracture
Aseptic
years
Other
Lysis
Peri-
Events at risk
Cuff
Acute trauma N (x100)
0.86 0.11 0.27 0.24 0.07 0.05 0.09
All cases 141 164.3
(0.73-1.01) (0.07-0.17) (0.20-0.36) (0.17-0.32) (0.04-0.12) (0.02-0.10) (0.06-0.15)
Proximal humeral 1.28 0.15 0.23 0.56 0.05 0.02 0.19
83 64.6
hemiarthroplasty (1.04-1.59) (0.08-0.29) (0.14-0.38) (0.40-0.77) (0.01-0.14) (0.00-0.11) (0.11-0.33)
Total shoulder
0 0.6 0 0 0 0 0 0 0
replacement
Reverse polarity
0.53 0.08 0.27 0.07 0.05 0.02
total shoulder 45 85.0 0
(0.40-0.71) (0.04-0.17) (0.18-0.41) (0.03-0.16) (0.02-0.13) (0.01-0.09)
replacement
0.93 0.07 0.43 0.21 0.14 0.21 0.07
Unconfirmed 13 14.0
(0.54-1.60) (0.01-0.51) (0.19-0.96) (0.07-0.67) (0.04-0.57) (0.07-0.67) (0.01-0.51)
Table 3.S10 and Table 3.S11 describe the prosthesis Cuff insufficiency is the leading indication for
time incidence rate (PTIR) per 100 years of follow-up revision for those who receive a proximal humeral
for the reported indication for revision in acute trauma hemiarthroplasty, whereas instability, dislocation, or
patients receiving a primary shoulder replacement. infection are the leading causes in reverse polarity
total shoulder replacements, see Table 3.S10. The
Table 3.S10 reports indications for all patients across low number of primary replacements and even lower
the life of the registry i.e. between 2012 and 2020, this frequency of revisions for patients whose data were
was achieved by aggregating indications for revision entered using the most recent minimum dataset
across the different minimum datasets. Table 3.S11 makes results difficult to interpret. It is important to
reports data for patients whose information was note that the indications for revision are not mutually
entered following the introduction of MDSv7. exclusive and 18.4%, 69.5%, and 9.9% recorded
none, one and two indications for revision respectively.
Table 3.S11 PTIR estimates of indications for shoulder revision (95% CI) for acute trauma by type of shoulder replacement using reports from MDSv7.
Prosthesis-
years
Events at risk
All causes
Aseptic
loosening
humerus
Aseptic
loosening
glenoid
Stiffness
Component
dissociation
Native
glenoid
surface
erosion
Implant
fracture
Dislocation
Unexplained
pain
Acute trauma N (x100)

1.31 0.11 0.04 0.07 0.04 0.04 0.04 0.34 0.04
All cases 35 26.7
(0.94-1.83) (0.04-0.35) (0.01-0.27) (0.02-0.30) (0.01-0.27) (0.01-0.27) (0.01-0.27) (0.18-0.65) (0.01-0.27)
12 5.6 0 0
Total shoulder
0 0.0 0 0 0 0 0 0 0 0 0
replacement
Reverse polarity
0.98 0.05 0.05 0.05 0.33

total shoulder 18 18.4 0 0 0 0
(0.62-1.56) (0.01-0.39) (0.01-0.39) (0.01-0.39) (0.15-0.73)
replacement
1.85 0.37
Unconfirmed 5 2.7 0 0 0 0 0 0 0
(0.77-4.45) (0.05-2.63)
Note: These have been suppressed due to zero events: impingement, glenoid implant wear, lysis humerus, lysis glenoid.
285
Table 3.S12 PTIR estimates of indications for shoulder revision (95% CI) for elective procedures by type of
shoulder replacement between 2012 and 2020.
insufficiency
Instablility |
Dislocation
indications
loosening |
All causes
prosthetic
Prosthesis-
Infection
fracture
Aseptic
years
Other
Lysis
Peri-
Events at risk
Cuff
Elective N (x100)
0.94 0.13 0.24 0.23 0.12 0.05 0.11
All cases 1,597 1,707.30
(0.89-0.98) (0.11-0.14) (0.22-0.27) (0.21-0.25) (0.10-0.13) (0.04-0.06) (0.10-0.13)
446 302.9
1.52 0.07 0.09 0.54 0.11 0.03 0.36
Resurfacing 229 150.8
(1.33-1.73) (0.04-0.12) (0.05-0.16) (0.44-0.68) (0.07-0.18) (0.01-0.07) (0.27-0.47)
1.70 0.04 0.08 0.53 0.08 0.53
Stemless 84 49.3 0
(1.38-2.11) (0.01-0.16) (0.03-0.22) (0.36-0.77) (0.03-0.22) (0.36-0.77)
1.29 0.10 0.16 0.41 0.10 0.33
Stemmed 133 102.9 0
(1.09-1.53) (0.05-0.18) (0.10-0.25) (0.30-0.55) (0.05-0.18) (0.24-0.46)

Total shoulder 0.74 0.06 0.25 0.36 0.11 0.02 0.08
399 540.0
replacement (0.67-0.81) (0.04-0.08) (0.21-0.30) (0.31-0.41) (0.09-0.15) (0.01-0.04) (0.06-0.11)
0.62 0.04 0.12 0.39 0.04 0.04
Resurfacing 16 25.8 0
(0.38-1.01) (0.01-0.28) (0.04-0.36) (0.21-0.72) (0.01-0.28) (0.01-0.28)
0.64 0.05 0.24 0.32 0.07 0.02 0.09
Stemless 108 167.7
(0.53-0.78) (0.03-0.10) (0.18-0.33) (0.24-0.41) (0.04-0.12) (0.01-0.06) (0.05-0.15)
0.79 0.06 0.26 0.38 0.14 0.02 0.09
Stemmed 275 346.6
(0.70-0.89) (0.04-0.09) (0.21-0.32) (0.32-0.45) (0.11-0.19) (0.01-0.05) (0.06-0.12)
Reverse polarity total 0.78 0.20 0.28 0.01 0.11 0.07 0.03
537 685.3
shoulder replacement (0.72-0.85) (0.17-0.24) (0.24-0.32) (0.01-0.03) (0.08-0.13) (0.05-0.09) (0.02-0.05)
2.09 0.32 0.32 0.16 0.80 0.16
Stemless 13 6.2 0
(1.22-3.60) (0.08-1.29) (0.08-1.29) (0.02-1.14) (0.34-1.93) (0.02-1.14)
0.77 0.20 0.28 0.01 0.10 0.07 0.03
Stemmed 524 679.1
(0.71-0.84) (0.17-0.24) (0.24-0.32) (0.01-0.03) (0.08-0.13) (0.05-0.09) (0.02-0.05)
Interpositional
0 0.1 0 0 0 0 0 0 0
arthroplasty
1.20 0.14 0.29 0.20 0.18 0.09 0.07
Unconfirmed 215 178.9
(1.05-1.37) (0.09-0.21) (0.22-0.38) (0.15-0.28) (0.13-0.26) (0.05-0.15) (0.04-0.12)
1.30 0.16 0.08 0.32 0.16 0.16 0.08
Unconfirmed HHA 16 12.3
(0.79-2.12) (0.04-0.65) (0.01-0.58) (0.12-0.86) (0.04-0.65) (0.04-0.65) (0.01-0.58)
1.26 0.06 0.23 0.32 0.20 0.05 0.10
Unconfirmed TSR 111 87.8
(1.05-1.52) (0.02-0.14) (0.15-0.35) (0.22-0.46) (0.13-0.33) (0.02-0.12) (0.05-0.20)
1.12 0.23 0.39 0.05 0.17 0.13 0.03
Unconfirmed RTSR 88 78.6
(0.91-1.38) (0.14-0.36) (0.28-0.56) (0.02-0.14) (0.10-0.28) (0.07-0.24) (0.01-0.10)
Unconfirmed IPA 0 0.1 0 0 0 0 0 0 0
Table 3.S13 PTIR estimates of indications for shoulder revision (95% CI) for elective procedures by type of shoulder replacement using reports from MDSv7.
Prosthesis-
years
Events at risk
All causes
Aseptic
loosening
humerus
Aseptic
loosening
glenoid
Stiffness
Impingement
Component
dissociation
Glenoid
implant
wear
Native glenoid
surface
erosion
Implant
fracture
Dislocation
Unexplained
pain
Elective N (x100)
10.09 0.23 0.79 0.23 0.19 0.98 0.23 0.33 0.05 1.72 0.51
All cases 217 21.5
(8.83-11.52) (0.10-0.56) (0.49-1.27) (0.10-0.56) (0.07-0.50) (0.64-1.50) (0.10-0.56) (0.16-0.68) (0.01-0.33) (1.25-2.37) (0.28-0.92)
Proximal humeral 11.41 0.60 1.80 0.60 0.60 4.20 3.00 2.40
19 1.7 0 0 0
hemiarthroplasty (7.28-17.89) (0.08-4.26) (0.58-5.59) (0.08-4.26) (0.08-4.26) (2.00-8.82) (1.25-7.21) (0.90-6.40)
11.19 4.48 2.24 4.48 6.71
Resurfacing 5 0.4 0 0 0 0 0 0
(4.66-26.88) (1.12-17.89) (0.32-15.88) (1.12-17.89) (2.16-20.81)
7.13 1.78 1.78
Stemless 4 0.6 0 0 0 0 0 0 0 0
(2.68-19.00) (0.25-12.66) (0.25-12.66)
15.21 1.52 1.52 1.52 6.08 7.60
Stemmed 10 0.7 0 0 0 0 0
(8.18-28.26) (0.21-10.80) (0.21-10.80) (0.21-10.80) (2.28-16.21) (3.16-18.27)
Total shoulder 6.00 0.65 0.32 0.32 0.49 0.49 1.14 0.32
37 6.2 0 0 0
replacement (4.35-8.29) (0.24-1.73) (0.08-1.30) (0.08-1.30) (0.16-1.51) (0.16-1.51) (0.54-2.38) (0.08-1.30)
Resurfacing 0 0.1 0 0 0 0 0 0 0 0 0 0 0
5.82 0.34 0.69 0.34 0.69 0.34
Stemless 17 2.9 0 0 0 0 0
(3.62-9.37) (0.05-2.43) (0.17-2.74) (0.05-2.43) (0.17-2.74) (0.05-2.43)
6.30 0.95 0.63 0.63 0.32 0.63 1.58 0.32
Stemmed 20 3.2 0 0 0
(4.07-9.77) (0.30-2.93) (0.16-2.52) (0.16-2.52) (0.04-2.24) (0.16-2.52) (0.66-3.79) (0.04-2.24)
Reverse polarity total 10.98 0.25 0.92 0.17 1.08 0.08 0.08 1.83 0.42
132 12.0 0 0
shoulder replacement (9.26-13.03) (0.08-0.77) (0.51-1.65) (0.04-0.67) (0.63-1.86) (0.01-0.59) (0.01-0.59) (1.21-2.78) (0.17-1.00)
22.84 11.42 11.42
Stemless <4 0.1 0 0 0 0 0 0 0 0
(5.71-91.33) (1.61-81.08) (1.61-81.08)
10.90 0.17 0.84 0.17 1.09 0.08 0.08 1.84 0.42
Stemmed 130 11.9 0 0
(9.18-12.94) (0.04-0.67) (0.45-1.56) (0.04-0.67) (0.63-1.88) (0.01-0.60) (0.01-0.60) (1.21-2.80) (0.17-1.01)
Interpositional
0 0.0 0 0 0 0 0 0 0 0 0 0 0
arthroplasty
17.46 0.60 1.20 2.41 1.81
Unconfirmed 29 1.7 0 0 0 0 0 0
(12.14-25.13) (0.08-4.27) (0.30-4.82) (0.90-6.42) (0.58-5.60)
8.03
Unconfirmed HHA <4 0.1 0 0 0 0 0 0 0 0 0 0
(1.13-57.03)
7.66 2.55
Unconfirmed TSR <4 0.4 0 0 0 0 0 0 0 0 0
(2.47-23.75) (0.36-18.12)
21.98 0.88 1.76 2.64 2.64
Unconfirmed RTSR 25 1.1 0 0 0 0 0 0
(14.85-32.53) (0.12-6.24) (0.44-7.03) (0.85-8.18) (0.85-8.18)
Unconfirmed IPA 0 0.0 0 0 0 0 0 0 0 0 0 0 0
Note: These have been suppressed due to zero events: lysis humerus, lysis glenoid.
287
Table 3.S12 and Table 3.S13 on the previous pages purposes of the annual report, revision procedures
describe the prosthesis time incidence rate (PTIR) include any addition, removal or modification of the
per 100 years of follow-up for the reported indication implants and procedures such as debridement and
for revision in elective patients receiving a primary implant retention with or without implant exchange,
shoulder replacement by type and sub-type of excision arthroplasty, amputation and conversion
shoulder replacement. to arthrodesis. For the avoidance of confusion,
completing a revision MDS form is also mandatory
Table 3.S12 reports indications for all patients across for a procedure involving modification of a joint by
the life of the registry i.e. between 2012 and 2020. adding another implant to another part of the joint.
This was achieved by aggregating indications for For the analyses of surgeon performance, hospital
revision across the different minimum datasets. Table performance and implant performance, debridement
3.S13 reports data for patients whose information was and implant retention without implant exchange is
entered following the introduction of MDSv7. currently excluded.
We have shown that cuff insufficiency is the leading
indication for revision for those who receive a
3.6.3 Patient Reported Outcome
proximal humeral hemiarthroplasty or conventional Measures (PROMs) Oxford Shoulder
total shoulder replacement, whereas instability or Scores (OSS) associated with primary
dislocation is the leading cause in reverse polarity shoulder replacement surgery
total shoulder replacements, see Table 3.S12. The
low number of primary replacements and even lower The OSS is a validated patient reported outcome
frequency of revisions for patients whose data were measure for use in shoulder surgery. It consists of
entered using the most recent minimum dataset 12 pain and function items which address problems
makes results difficult to interpret. It is important that the patient may have encountered with their
to note the indications for revision are not mutually shoulder over the preceding four weeks (Dawson et
exclusive and 22.2%, 64.7%, and 11.1% recorded al., 1996). The score is coded from 0 to 4 (from ‘worst’
none, one and two indications for revision respectively. to ‘best’) and then summed in line with updated OSS
recommendations (Dawson et al., 2009). The final total
The NJR asks surgeons and those responsible for score ranges from 0 to 48, with 48 representing the
healthcare delivery to ensure that when primary and ‘best’ outcome and 0 the ‘worst’. Where up to two
revision joint replacement procedures of the hip, items were missing, the average of the remaining items
knee, ankle, elbow or shoulder are performed, that can be substituted for the missing values (Dawson et
the relevant MDS form is completed and data entered al., 2009). If more than two items were missing, the
into the registry. This is a requirement mandated by results have to be disregarded.
the Department of Health and Social Care. For the
Dawson J, Fitzpatrick R, Carr A, JBJS, 1996: 78-B, 593-600.

Dawson J, Rogers K, Fitzpatrick R and Carr A, Arch Orthop Trauma Surg, 2009, 129:119-123.
Table 3.S14 Number and percentage of patients who completed an Oxford Shoulder Score by acute trauma and elective indications, by the
collection window of interest at different time points.
Pre-operative OSS 6 Month OSS 3 Year OSS 5 Year OSS
Eligible Responders Eligible Responders Eligible Responders Eligible Responders

N (%) (%) N (%) (%) N (%) (%) N (%) (%)
All eligible cases 5,143 (100.0) 4,725 (100.0) 2,510 (100.0) 1,178 (100.0)
All responders 487 (9.5) (100.0) 2,103 (44.5) (100.0) 256 (10.2) (100.0) 340 (28.9) (100.0)
All complete* within window of interest 347 (6.7) (71.3) 1,539 (32.6) (73.2) 249 (9.9) (97.3) 335 (28.4) (98.5)
OSS collected before window of interest <4 (<0.1) (0.2) 5 (0.1) (0.2) <4 (<0.1) (0.4) <4 (0.1) (0.3)
1 to 9 Items completed <4 (<0.1) (0.2) 0 (0) (0) 0 (0) (0) 0 (<0.1) (<0.1)
10 to 11 Items completed 0 (0) (0) <4 (<0.1) (<0.1) 0 (<0.1) (<0.1) <4 (0.1) (0.3)
12 Items completed 0 (0) (0) 4 (0.1) (0.2) <4 (<0.1) (0.4) 0 (0) (0)
OSS collected within window of interest 357 (6.9) (73.3) 1,550 (32.8) (73.7) 253 (10.1) (98.8) 337 (28.6) (99.1)
1 to 9 Items completed 10 (0.2) (2.1) 11 (0.2) (0.5) 4 (0.2) (1.6) <4 (0.2) (0.6)
Acute trauma
10 to 11 Items completed 22 (0.4) (4.5) 85 (1.8) (4.0) 6 (0.2) (2.3) 10 (0.8) (2.9)
12 Items completed 325 (6.3) (66.7) 1,454 (30.8) (69.1) 243 (9.7) (94.9) 325 (27.6) (95.6)
OSS collected after window of interest 129 (2.5) (26.5) 525 (11.1) (25.0) <4 (0.1) (0.8) <4 (0.2) (0.6)
1 to 9 Items completed 19 (0.4) (3.9) <4 (0.1) (0.1) 0 (<0.1) (<0.1) 0 (<0.1) (<0.1)
10 to 11 Items completed 8 (0.2) (1.6) 35 (0.7) (1.7) 0 (<0.1) (<0.1) 0 (<0.1) (<0.1)
12 Items completed 102 (2.0) (20.9) 487 (10.3) (23.2) <4 (0.1) (0.8) <4 (0.2) (0.6)
All eligible cases 45,112 (100.0) 43,145 (100.0) 27,045 (100.0) 14,417 (100.0)
All responders 16,940 (37.6) (100.0) 21,288 (49.3) (100.0) 3,045 (11.3) (100.0) 4,409 (30.6) (100.0)
All complete* within window of interest 11,405 (25.3) (67.3) 14,992 (34.7) (70.4) 2,989 (11.1) (98.2) 4,333 (30.1) (98.3)
OSS collected before window of interest 1,006 (2.2) (5.9) 110 (0.3) (0.5) 7 (<0.1) (0.2) 15 (0.1) (0.3)
1 to 9 Items completed 5 (<0.1) (<0.1) <4 (<0.1) (<0.1) 0 (<0.1) (<0.1) 0 (<0.1) (<0.1)
10 to 11 Items completed 25 (0.1) (0.1) 10 (<0.1) (<0.1) <4 (<0.1) (<0.1) <4 (<0.1) (<0.1)
12 Items completed 976 (2.2) (5.8) 99 (0.2) (0.5) 6 (<0.1) (0.2) 14 (0.1) (0.3)
OSS collected within window of interest 11,462 (25.4) (67.7) 15,093 (35.0) (70.9) 3,018 (11.2) (99.1) 4,382 (30.4) (99.4)
Elective
12 Items completed 10,947 (24.3) (64.6) 14,136 (32.8) (66.4) 2,777 (10.3) (91.2) 4,045 (28.1) (91.7)
OSS collected after window of interest 4,472 (9.9) (26.4) 5,892 (13.7) (27.7) 20 (0.1) (0.7) 12 (0.1) (0.3)
1 to 9 Items completed 130 (0.3) (0.8) 38 (0.1) (0.2) 0 (<0.1) (<0.1) <4 (<0.1) (<0.1)
10 to 11 Items completed 375 (0.8) (2.2) 300 (0.7) (1.4) 0 (<0.1) (<0.1) <4 (<0.1) (0.1)
12 Items completed 3,967 (8.8) (23.4) 5,554 (12.9) (26.1) 20 (0.1) (0.7) 8 (0.1) (0.2)
*Complete corresponds to ten or more items completed.

Note: The windows of interest are: Pre-operative[-90 to 0 days], 6 months [5 to 8 months], 3 years [2 years 11 months to 3 years 6 months], 5 years [4 years 11 months to 5 years 6 months].
289
Table 3.S14 provides a detailed description of the of a surgical technique or implantable construct,
number of patients reporting an OSS pre-operatively, understanding where the patient started is critical
6 months, 3 years and 5 years following surgery for in order to understand how the patient is likely to
patients undergoing primary shoulder replacement for respond to surgery. Collecting a pre-operative PROM
acute trauma or elective indications. The responses post-operatively is likely to induce recall bias and for
are further divided by how close to the time point of this reason the end of the pre-operative window was
interest it was collected and the completeness of each strictly defined as the day of surgery. Table 3.S14
PROM. The results are expressed absolutely (N) and clearly illustrates only a small minority of eligible
as a percentage (%) of ‘Eligible’ participants and those patients complete an OSS questionnaire prior to
who ‘Responded’ to the PROMs. Eligibility is defined as surgery and within the window of interest.
being alive at the time point of interest and also having
sufficient follow-up time following primary surgery. Given the low compliance in pre-operative score
collection by hospitals delivering shoulder replacement
How close the response was to the time point of surgery, the potential for bias in interpreting results
interest is categorised by defining ‘windows of is clear. Collection and compliance with reporting
interest’. The pre-operative window of interest is at 6 months, 3 and 5 years is substantially better
90 days prior to the primary surgery until the day of than pre-operative rates, but the response rate of
the primary operation. The 6-month data collection all eligible participants is still less than 50% in all
window of interest ranges from 5 months to 8 instances. The British Elbow and Shoulder Society
months, i.e. spanning a 3-month window of interest. (BESS) have deemed shoulder PROMs essential in the
The 3 and 5 year data collections had windows of assessment of patient outcomes and surveillance after
interest ranging from 1 month prior to 3 and 5 years shoulder replacement surgery. The low pre-operative
respectively to 6 months after i.e. spanning a 7-month compliance with PROMs data collection by hospital
window of interest. trusts is therefore particularly concerning.
Ensuring data is collected pre-operatively by hospital

trusts is very important. In order to assess the efficacy
Table 3.S15 Number and percentage of patients who completed cross-sectional Oxford Shoulder Score by
overall, acute trauma, elective and by year of primary operation, within the collection window of interest, with
valid measurements at the time points of interest.
Oxford Shoulder Scores completed at:

Potential
Pre-op 6 month 3 year 5 year
Year of cases
primary N N (% of Pre-op) N (% of Pre-op) N (% of Pre-op) N (% of Pre-op)
All years 50,255 11,752 (23.4) 16,531 (34.5) 3,238 (11.0) 4,668 (29.9)
Acute trauma & elective
2012 2,545 673 (26.4) 345 (13.7) 0 (0) 1,130 (51.5)

2013 4,412 1,077 (24.4) 1,883 (43.1) 0 (0) 1,354 (35.5)
2014 5,309 1,414 (26.6) 299 (5.7) 2,067 (41.7) 1,839 (39.9)
2015 5,734 1,489 (26.0) 857 (15.0) 729 (13.6) 345 (6.9)
2016 6,537 1,476 (22.6) 26 (0.4) 263 (4.3) 0
2017 7,002 1,486 (21.2) 4,672 (67.3) 179 (2.7) 0
2018 7,223 1,429 (19.8) 5,024 (70.1) 0 0

2019 7,660 1,781 (23.3) 3,360 (44.2) 0 0
2020 3,833 927 (24.2) 65 (3.6) 0 0
All years 5,143 347 (6.7) 1,539 (32.6) 249 (9.9) 335 (28.4)
2012 160 11 (6.9) 17 (11.0) 0 (0) 52 (40.9)
2013 387 42 (10.9) 149 (39.4) 0 (0) 100 (33.3)
Acute trauma
2014 473 36 (7.6) 33 (7.2) 162 (40.8) 145 (42.3)

2015 532 31 (5.8) 92 (17.7) 76 (16.4) 38 (9.3)
2016 598 41 (6.9) 7 (1.2) 9 (1.7) 0
2017 713 35 (4.9) 441 (63.4) <4 (0.3) 0
2018 768 50 (6.5) 471 (62.3) 0 0
2019 864 53 (6.1) 323 (38.2) 0 0
2020 648 48 (7.4) 6 (1.8) 0 0
All years 45,112 11,405 (25.3) 14,992 (34.7) 2,989 (11.1) 4,333 (30.1)
2012 2,385 662 (27.8) 328 (13.9) 0 (0) 1,078 (52.1)
2013 4,025 1,035 (25.7) 1,734 (43.4) 0 (0) 1,254 (35.7)
2014 4,836 1,378 (28.5) 266 (5.5) 1,905 (41.8) 1,694 (39.7)
Elective
2015 5,202 1,458 (28.0) 765 (14.8) 653 (13.3) 307 (6.7)
2016 5,939 1,435 (24.2) 19 (0.3) 254 (4.5) 0
2017 6,289 1,451 (23.1) 4,231 (67.8) 177 (3.0) 0
2018 6,455 1,379 (21.4) 4,553 (71.0) 0 0
2019 6,796 1,728 (25.4) 3,037 (45.0) 0 0
2020 3,185 879 (27.6) 59 (4.0) 0 0
Table 3.S15 provides a detailed description of the appear without a percentage in parentheses, the
number of patients reporting complete OSS within the PROMs were collected prior to the target date but
window of interest pre-operatively and at 6 months, 3 within the window of interest. The data illustrates that
years and 5 years by the year of surgery for patients collection and submission of pre-operative PROMs by
undergoing primary shoulder replacement for acute hospitals is consistently poor, with less than 30% of
trauma or elective indications. The denominator used elective patients having their PROMs data submitted.
to calculate percentages is the number of patients In recent years the compliance with 6-month reporting
alive at the milestone of interest. Where numbers has steadily improved.
Table 3.S16 Number and percentage of patients who completed longitudinal Oxford Shoulder Score by overall,
acute trauma, elective and by year of primary operation, within the collection window of interest, with valid
measurements at the time points of interest.
Oxford Shoulder Scores completed at:

Pre-op,
Potential Pre-op Pre-op, 6m Pre-op, 3y Pre-op, 5y Pre-op, 6m, 3y 6m, 3y, 5y
Year of cases
primary N N N (% of Pre-op) N (% of Pre-op) N (% of Pre-op) N (% of Pre-op) N (% of Pre-op)
All years 50,255 11,752 3,863 (32.9) 1,141 (9.7) 1,372 (11.7) 355 (3.0) 118 (1.0)
2012 2,545 673 92 (13.7) 0 (0) 345 (51.3) 0 (0) 0 (0)
2013 4,412 1,077 527 (48.9) 0 (0) 369 (34.3) 0 (0) 0 (0)
2014 5,309 1,414 83 (5.9) 614 (43.4) 561 (39.7) 62 (4.4) 49 (3.5)
2015 5,734 1,489 239 (16.1) 201 (13.5) 97 (6.5) 185 (12.4) 69 (4.6)
2016 6,537 1,476 5 (0.3) 197 (13.3) 0 (0) <4 (0.2) 0 (0)
2017 7,002 1,486 1,048 (70.5) 129 (8.7) 0 (0) 105 (7.1) 0 (0)
2018 7,223 1,429 1,054 (73.8) 0 (0) 0 (0) 0 (0) 0 (0)
2019 7,660 1,781 762 (42.8) 0 (0) 0 (0) 0 (0) 0 (0)

2020 3,833 927 53 (5.7) 0 (0) 0 (0) 0 (0) 0 (0)
All years 5,143 347 102 (29.4) 25 (7.2) 29 (8.4) <4 (0.6) <4 (0.3)
2012 160 11 <4 (9.1) 0 (0) 4 (36.4) 0 (0) 0 (0)
2013 387 42 17 (40.5) 0 (0) 13 (31.0) 0 (0) 0 (0)
Acute trauma
2014 473 36 <4 (2.8) 14 (38.9) 11 (30.6) 0 (0) 0 (0)

2015 532 31 <4 (9.7) <4 (6.5) <4 (3.2) <4 (3.2) <4 (3.2)
2016 598 41 0 (0) 7 (17.1) 0 (0) 0 (0) 0 (0)
2017 713 35 21 (60.0) <4 (5.7) 0 (0) <4 (2.9) 0 (0)
2018 768 50 33 (66.0) 0 (0) 0 (0) 0 (0) 0 (0)
2019 864 53 20 (37.7) 0 (0) 0 (0) 0 (0) 0 (0)
2020 648 48 6 (12.5) 0 (0) 0 (0) 0 (0) 0 (0)
All years 45,112 11,405 3,761 (33.0) 1,116 (9.8) 1,343 (11.8) 353 (3.1) 117 (1.0)
2012 2,385 662 91 (13.7) 0 (0) 341 (51.5) 0 (0) 0 (0)
2013 4,025 1,035 510 (49.3) 0 (0) 356 (34.4) 0 (0) 0 (0)
2014 4,836 1,378 82 (6.0) 600 (43.5) 550 (39.9) 62 (4.5) 49 (3.6)
Elective
2015 5,202 1,458 236 (16.2) 199 (13.6) 96 (6.6) 184 (12.6) 68 (4.7)
2016 5,939 1,435 5 (0.3) 190 (13.2) 0 (0) <4 (0.2) 0 (0)
2017 6,289 1,451 1,027 (70.8) 127 (8.8) 0 (0) 104 (7.2) 0 (0)
2018 6,455 1,379 1,021 (74.0) 0 (0) 0 (0) 0 (0) 0 (0)
2019 6,796 1,728 742 (42.9) 0 (0) 0 (0) 0 (0) 0 (0)
2020 3,185 879 47 (5.3) 0 (0) 0 (0) 0 (0) 0 (0)
Table 3.S16 describes the number and percentage with no more than two items missing responses. The
of paired measurements available for longitudinal proportion of patients available for a paired longitudinal
analyses for all patients undergoing primary shoulder analysis at any time point is low, and the proportion of
replacement for acute trauma or elective indications. patients with serial measurements at any time point is
The denominator used to calculate percentages is even lower. While the proportion of patients with pre-
the number of pre-operative measurements. The operative and 6-month OSS has increased in recent
numerator is the number of responses within the years, this still only represents 14.6% of all eligible
window of interest, see Table 3.S14 on page 289, primary replacements.
Figure 3.S7 KM estimates of cumulative revision for primary elective shoulder replacements for patients
with and without valid PROMs. Blue italics in the numbers at risk table signify that fewer than 250 cases
10

0
0 1 2 3 4 5 6 7 8
Years since primary
Non PROMs cohort 41,351 37,117 30,304 24,279 18,722 12,912 8,341 4,285 1,569
PROMs cohort 3,761 3,669 2,865 1,841 831 799 567 465 72
Figure 3.S7 reports the cumulative revision rate who are not responding and so are not representative
for elective patients undergoing primary shoulder of the larger population. This highlights the risk of
replacements who completed pre-operative and using incomplete datasets to make inferences for
6-month PROMs assessments within the specified the larger cohort and this PROMs data needs to be
window of interest. Results indicate a different interpreted cautiously despite its relatively large size. If
cumulative revision rate for patients who are included anything it indicates that the PROMs cohort is likely to
in the PROMs cohort versus those who are not. This be a more ‘satisfied’ group of patients as their revision
difference suggests the group of patients responding rates are lower than the non-PROMs cohort.
to the PROMs questionnaires are different from those
Figure 3.S8 Distribution and scatter of pre-operative OSS and the change in OSS (post-pre) score
for those receiving elective shoulder replacements for valid measurements within the collection window
of interest.
Frequency Pre−operative OSS

x10 [0m]
30
20
10
10
20
30
40
48
0
48 48
40 40
30 30
20 20
Difference in OSS
Difference in OSS
10 10
[6m−0m]
[6m−0m]
0 0
−10 −10
−20 −20
−30 −30
−40 −40
−48 −48
Frequency
0
x10
10
20
0
10
20
30
40
48
Pre−operative OSS [0m]
Note: Elective patients with pre and 6 month presented only, N=3,761.
Figure 3.S8 illustrates the distribution of pre-operative the necessity of ascertaining a pre-operative PROMs
OSS and change in OSS between the pre-operative when assessing the efficacy of any intervention
and the 6-month assessment. Results are displayed associated with a primary shoulder replacement. In the
for patients with elective indications for primary absence of specialist methods which account for floor
shoulder replacement only. It also illustrates the and ceiling effects, a simple analysis of change scores
association between pre-operative OSS and the is reported to be the most appropriate (Glymour et al.,
change in OSS. While pre-operative and change in 2005). At 6 months following surgery, 5.3% of patients
OSS are approximately normally distributed, this hides reported a score worse than they did pre-operatively.
the profound ceiling effect within the assessment of This figure is reduced compared to previous years due
the change score. This makes the interpretation of to the more refined inclusion/exclusion criteria of the
change in OSS particularly challenging and highlights PROMs cohort as defined previously.
Glymour M., et al. American Journal of Epidemiology, 2005: 162(3), 267-278.
Table 3.S17 Descriptive statistics of the pre-operative, 6-month and the change in OSS by overall,
acute trauma, elective and by year of primary operation, within the collection window of interest, with valid
measurements pre-operatively and 6 months post-operatively.
Oxford Shoulder Score [0 min, 48 max]

Complete Pre-op 6 month (6 month - Pre-op)
Year of cases
primary N Mean (SD) [25,50,75]th Mean (SD) [25,50,75]th Mean (SD) [25,50,75]th
All years 3,863 16.7 (8.4) [11, 16, 22] 35.7 (10.4) [30, 39, 44] 19.0 (11.6) [12, 20, 27]
2012 92 17.6 (7.9) [12, 16, 23] 33.8 (11.7) [28, 37, 43] 16.2 (11.7) [9, 16, 25]
2013 527 17.5 (8.6) [11, 17, 23] 33.8 (10.7) [27, 36, 43] 16.3 (12.0) [8, 17, 25]
2014 83 16.2 (8.0) [10, 15, 22] 34.0 (11.1) [25, 36, 42] 17.7 (10.2) [12, 17, 25]
2015 239 16.0 (7.7) [11, 15, 21] 33.8 (11.1) [28, 36, 43] 17.8 (11.0) [10, 19, 26]
2016 5 17.4 (9.3) [9, 18, 26] 42.6 (6.1) [37, 46, 47] 25.2 (11.4) [22, 28, 29]
2017 1,048 16.8 (8.4) [11, 16, 22] 36.0 (10.2) [30, 39, 44] 19.2 (11.6) [12, 20, 28]
2018 1,054 16.4 (8.5) [10, 16, 22] 36.2 (10.4) [30, 39, 44] 19.8 (11.7) [13, 21, 28]
2019 762 16.8 (8.4) [11, 16, 22] 36.7 (10.0) [31, 40, 44] 19.9 (11.1) [13, 21, 28]

2020 53 15.4 (9.5) [9, 14, 24] 36.9 (9.0) [31, 40, 44] 21.5 (9.5) [17, 22, 28]
All years 102 13.5 (15.9) [1, 8, 23] 31.4 (11.7) [22, 34, 42] 17.8 (20.3) [6, 22, 33]
2012 <4
2013 17 11.9 (14.7) [2, 8, 12] 33.3 (13.8) [25, 41, 44] 21.3 (23.8) [17, 27, 40]
Acute trauma
2014 <4
2015 <4
2016 0
2017 21 15.4 (17.3) [1, 8, 24] 31.4 (10.7) [22, 34, 36] 16.0 (21.3) [3, 22, 27]
2018 33 16.6 (16.2) [4, 11, 28] 28.7 (10.8) [20, 30, 37] 12.1 (19.7) [-3, 14, 29]
2019 20 9.9 (15.6) [0, 1, 15] 33.7 (12.7) [25, 40, 44] 23.8 (17.7) [15, 26, 39]
2020 6 6.3 (15.0) [0, 0, 1] 33.1 (11.3) [22, 35, 43] 26.8 (14.2) [20, 25, 42]
All years 3,761 16.8 (8.1) [11, 16, 22] 35.8 (10.4) [30, 39, 44] 19.0 (11.3) [12, 20, 27]
2012 91 17.4 (7.5) [12, 16, 23] 33.8 (11.7) [28, 36, 43] 16.4 (11.6) [9, 16, 25]
2013 510 17.7 (8.3) [11, 17, 23] 33.8 (10.6) [27, 36, 43] 16.2 (11.4) [8, 17, 24]
2014 82 16.3 (8.0) [10, 15, 22] 34.2 (10.9) [26, 37, 42] 17.9 (10.2) [12, 17, 25]
Elective
2015 236 16.1 (7.6) [11, 16, 21] 33.8 (11.1) [28, 36, 43] 17.7 (10.9) [10, 19, 26]
2016 5 17.4 (9.3) [9, 18, 26] 42.6 (6.1) [37, 46, 47] 25.2 (11.4) [22, 28, 29]
2017 1,027 16.8 (8.2) [11, 16, 22] 36.1 (10.2) [30, 39, 44] 19.2 (11.4) [12, 20, 28]
2018 1,021 16.4 (8.2) [11, 16, 22] 36.4 (10.3) [31, 39, 44] 20.1 (11.3) [13, 21, 28]
2019 742 17.0 (8.0) [11, 16, 22] 36.8 (9.9) [31, 40, 44] 19.8 (10.8) [13, 21, 28]
2020 47 16.5 (8.0) [10, 15, 25] 37.4 (8.7) [31, 40, 44] 20.9 (8.8) [16, 21, 28]
Table 3.S17 presents descriptive statistics, mean and patients with valid OSS that receive primary shoulder
standard deviation, median and interquartile range, replacements is relatively low, however, the results
by year of primary shoulder replacements overall, and appear to be broadly concordant with those receiving
by those receiving shoulder replacements for acute primary shoulder replacement for elective indications.
trauma or elective indications. Results are presented The change in OSS has tended to improve across the
only for those with measurements pre-operatively life of the registry, but the significance of this is very
and at 6 months, within the window of interest and unclear given the potential for bias due to the lack of a
with no more than two items missing. The number of representative sample.
trauma, elective and by shoulder type, within the collection window of interest, with valid measurements pre-
operatively and 6 months post-operatively.

cases
Primary procedure N Mean (SD) [25,50,75]th Mean (SD) [25,50,75]th Mean (SD) [25,50,75]th
Proximal humeral hemiarthroplasty 404 17.8 (9.2) [11, 17, 23] 31.3 (11.8) [23, 34, 41] 13.5 (12.4) [6, 14, 23]
Resurfacing 185 18.4 (8.4) [12, 18, 23] 32.2 (11.3) [26, 35, 41] 13.8 (11.3) [7, 14, 23]
Stemless 80 19.9 (8.4) [16, 19, 23] 33.2 (11.3) [26, 36, 42] 13.4 (10.5) [6, 14, 21]
Stemmed 139 15.8 (10.2) [9, 14, 22] 29.1 (12.3) [18, 32, 39] 13.3 (14.8) [5, 14, 24]
Total shoulder replacement 1,237 17.6 (8.0) [12, 17, 23] 38.5 (9.2) [35, 41, 45] 20.9 (10.5) [14, 22, 29]
Resurfacing 56 18.6 (8.2) [12, 18, 25] 39.1 (7.1) [36, 40, 45] 20.5 (9.3) [14, 21, 26]
Stemless 547 18.0 (8.1) [12, 17, 24] 38.9 (9.0) [36, 41, 45] 20.9 (10.3) [15, 22, 29]
Stemmed 634 17.2 (8.0) [11, 17, 23] 38.1 (9.6) [34, 41, 45] 20.9 (10.8) [14, 21, 29]
Reverse polarity total shoulder
1,974 15.9 (8.4) [10, 15, 21] 34.9 (10.5) [29, 37, 43] 19.0 (11.7) [12, 20, 27]
replacement
Stemless 31 16.9 (7.1) [9, 17, 23] 36.4 (10.1) [28, 40, 45] 19.5 (12.3) [7, 21, 29]
Stemmed 1,943 15.9 (8.4) [10, 15, 21] 34.9 (10.5) [29, 37, 43] 19.0 (11.7) [12, 20, 27]
Interpositional arthroplasty 0
Unconfirmed 248 16.8 (8.9) [10, 16, 24] 34.6 (10.3) [28, 37, 43] 17.7 (11.2) [11, 18, 25]
Unconfirmed HHA 13 17.0 (7.3) [11, 14, 23] 28.2 (14.1) [18, 29, 42] 11.2 (14.2) [4, 10, 21]
Unconfirmed TSR 114 17.3 (8.7) [10, 18, 24] 35.6 (10.6) [29, 39, 44] 18.3 (11.5) [11, 19, 27]
Unconfirmed RTSR 121 16.3 (9.2) [10, 16, 22] 34.2 (9.3) [28, 36, 41] 17.9 (10.5) [11, 18, 25]
Unconfirmed IPA 0
Resurfacing 0
Stemless 0
Stemmed 22 16.3 (17.5) [3, 10, 28] 28.5 (13.4) [18, 30, 41] 12.2 (25.8) [2, 17, 30]
Total shoulder replacement <4
Resurfacing 0
Stemless <4
Acute trauma
Stemmed 0
75 13.4 (15.8) [0, 8, 24] 32.1 (11.3) [24, 34, 42] 18.8 (18.7) [7, 22, 34]
replacement
Stemless 0
Stemmed 75 13.4 (15.8) [0, 8, 24] 32.1 (11.3) [24, 34, 42] 18.8 (18.7) [7, 22, 34]
Unconfirmed 4 1.8 (3.5) [0, 0, 4] 29.8 (9.6) [22, 30, 38] 28.0 (12.0) [18, 30, 38]
Unconfirmed HHA 0
Unconfirmed TSR 0
Unconfirmed IPA 0
Table 3.S18 (continued)

cases
Primary procedure N Mean (SD) [25,50,75]th Mean (SD) [25,50,75]th Mean (SD) [25,50,75]th
Resurfacing 185 18.4 (8.4) [12, 18, 23] 32.2 (11.3) [26, 35, 41] 13.8 (11.3) [7, 14, 23]
Stemless 80 19.9 (8.4) [16, 19, 23] 33.2 (11.3) [26, 36, 42] 13.4 (10.5) [6, 14, 21]
Stemmed 117 15.7 (8.3) [10, 14, 22] 29.2 (12.2) [21, 32, 39] 13.5 (11.8) [5, 14, 22]

Total shoulder replacement 1,236 17.6 (8.0) [12, 17, 23] 38.5 (9.2) [35, 41, 45] 20.9 (10.5) [14, 22, 29]
Resurfacing 56 18.6 (8.2) [12, 18, 25] 39.1 (7.1) [36, 40, 45] 20.5 (9.3) [14, 21, 26]
Stemless 546 18.0 (8.1) [12, 18, 24] 38.9 (9.0) [36, 41, 45] 20.9 (10.3) [15, 22, 29]
Stemmed 634 17.2 (8.0) [11, 17, 23] 38.1 (9.6) [34, 41, 45] 20.9 (10.8) [14, 21, 29]
Elective

1,899 16.0 (7.9) [10, 15, 21] 35.0 (10.4) [29, 37, 43] 19.0 (11.3) [12, 20, 27]
replacement
Stemless 31 16.9 (7.1) [9, 17, 23] 36.4 (10.1) [28, 40, 45] 19.5 (12.3) [7, 21, 29]
Stemmed 1,868 16.0 (8.0) [10, 15, 21] 35.0 (10.4) [29, 37, 43] 19.0 (11.3) [12, 20, 27]
Unconfirmed 244 17.1 (8.7) [10, 17, 24] 34.7 (10.3) [28, 37, 43] 17.6 (11.1) [10, 18, 25]
Unconfirmed HHA 13 17.0 (7.3) [11, 14, 23] 28.2 (14.1) [18, 29, 42] 11.2 (14.2) [4, 10, 21]
Unconfirmed TSR 114 17.3 (8.7) [10, 18, 24] 35.6 (10.6) [29, 39, 44] 18.3 (11.5) [11, 19, 27]
Unconfirmed IPA 0
Table 3.S18 presents descriptive statistics, mean and Table 3.S18 clearly illustrates that the change between
standard deviation, median and interquartile range, by pre-operative and 6-month assessment of OSS while
type and sub-type of primary shoulder replacements positive, is still substantially less for patients receiving a
overall, and by those receiving shoulder replacements proximal humeral hemiarthroplasty compared to either
for acute trauma or elective indications. Results are a conventional total or reverse polarity total shoulder
presented only for those with measurements pre- replacement. The change in OSS between conventional
operatively and at 6 months, within the window of total shoulder replacement versus reverse polarity total
interest and with no more than two items missing. shoulder replacement and sub-type versus type of
The number of patients receiving a primary shoulder shoulder replacement is broadly similar.
replacement for acute trauma indications is small.
3.6.4 Mortality after primary shoulder to eight years following the primary procedure for all
patients (Table 3.S19 only) and patients undergoing
replacement surgery surgery for acute trauma and elective indications
This following section describes the mortality profile separately. Data is shown at 30 and 90 days following
for patients receiving primary shoulder replacements. the primary procedure and then every year until the
Where patients received same-day bilateral procedures eighth year. Table 3.S19 indicates the importance of
(N=25), see Figure 3.S1 (page 259), they were excluded separating the data for patients receiving a primary
from the analysis to avoid double counting. This results shoulder replacement for acute trauma from the
in 45,765 patient procedures being included in the data for those with elective indications, due to the
analysis, with 5,512 observed deaths. differences in the frailty of the patient population,
despite their similar age profile, see Table 3.S2 on
Figure 3.S9 and Table 3.S19 describe the mortality of page 265.
patients receiving a primary shoulder replacement up
Figure 3.S9 KM estimates of cumulative mortality by acute trauma and elective indications for patients
undergoing primary shoulder replacement. Blue italics in the numbers at risk table signify that fewer than
250 cases remained at risk at these time points.
45
40
35
30
25
20
15
10
0
0 1 2 3 4 5 6 7 8
Years since primary
Acute trauma 5,131 4,307 3,344 2,504 1,775 1,175 710 367 93
Elective 45,099 41,342 33,989 27,037 20,411 14,411 9,447 5,099 1,786
Table 3.S19 KM estimates of cumulative mortality (95% CI) by acute trauma and elective indications for patients undergoing primary shoulder
replacement. Blue italics signify that fewer than 250 cases remained at risk at these time points.

primary Male
N Median (IQR) (%) 30 days 90 days 1 year 2 years 3 years 4 years 5 years 6 years 7 years 8 years
All 0.17 0.38 1.61 3.67 6.30 9.54 13.14 17.00 21.31 26.04
50,230 73 (67 to 79) 30
cases (0.14-0.21) (0.33-0.44) (1.50-1.73) (3.50-3.85) (6.06-6.54) (9.23-9.85) (12.76-13.54) (16.52-17.49) (20.70-21.93) (25.20-26.90)
Acute 0.69 1.30 3.93 7.68 11.97 17.00 23.03 28.85 34.64 40.31
5,131 74 (67 to 80) 23
trauma (0.49-0.95) (1.02-1.66) (3.42-4.51) (6.93-8.51) (10.98-13.03) (15.75-18.34) (21.45-24.70) (26.91-30.89) (32.24-37.17) (37.05-43.76)
0.12 0.28 1.36 3.24 5.70 8.76 12.14 15.83 20.00 24.67
Elective 45,099 73 (67 to 79) 31
(0.09-0.15) (0.23-0.33) (1.25-1.47) (3.07-3.42) (5.46-5.94) (8.45-9.08) (11.75-12.55) (15.34-16.33) (19.38-20.64) (23.81-25.56)
299
Figure 3.S10 KM estimates of cumulative mortality for primary elective shoulder replacement by gender.
30
25
20
15
10
0
0 1 2 3 4 5 6 7 8
Years since primary
Female 31,339 28,829 23,760 18,981 14,365 10,206 6,680 3,595 1,257
Male 13,760 12,513 10,229 8,056 6,046 4,205 2,767 1,504 529
Figure 3.S11 KM estimates of cumulative mortality for primary elective shoulder replacement by age
group and gender. Blue italics in the numbers at risk table signify that fewer than 250 cases remained at
risk at these time points.
Males Females
45 45
40 40
35 35

30 30

25 25
20 20
15 15
10 10
5 5
0 0
0 1 2 3 4 5 6 7 8 0 1 2 3 4 5 6 7 8
<55y 1,479 1,370 1,164 955 752 539 370 220 91 1,145 1,077 935 786 637 453 305 176 74
55 to 64y 2,613 2,371 1,948 1,557 1,198 857 583 330 106 3,263 2,997 2,473 2,020 1,574 1,190 802 476 187
65 to 74y 5,241 4,813 3,988 3,150 2,371 1,675 1,113 609 212 11,551 10,717 8,892 7,184 5,491 3,986 2,662 1,453 510
≥75y 4,427 3,959 3,129 2,394 1,725 1,134 701 345 120 15,380 14,038 11,460 8,991 6,663 4,577 2,911 1,490 486
302
Table 3.S20 KM estimates of cumulative mortality (95% CI) for primary shoulder replacement for elective cases by gender and age group.
primary
Gender (years) N 30 days 90 days 1 year 2 years 3 years 4 years 5 years 6 years 7 years 8 years
0.10 0.24 1.15 2.93 5.36 8.44 11.91 15.77 19.97 24.85
All 31,339
(0.07-0.14) (0.20-0.31) (1.04-1.28) (2.74-3.13) (5.09-5.65) (8.07-8.82) (11.44-12.39) (15.19-16.37) (19.23-20.74) (23.81-25.92)
0.18 0.63 1.72 2.29 4.06 5.61 6.11 8.69 9.74
<55 1,145 0
(0.04-0.70) (0.30-1.31) (1.08-2.71) (1.52-3.43) (2.93-5.62) (4.18-7.52) (4.56-8.16) (6.45-11.66) (7.19-13.12)
0.03 0.06 0.45 1.30 2.27 3.56 4.56 6.72 8.26 10.09
55 to 64 3,263
(0.00-0.22) (0.02-0.25) (0.26-0.75) (0.95-1.78) (1.76-2.92) (2.87-4.41) (3.73-5.58) (5.57-8.10) (6.85-9.94) (8.22-12.36)
Female
0.08 0.18 0.75 2.00 3.75 5.65 8.00 10.43 13.21 16.49
65 to 74 11,551
(0.04-0.15) (0.12-0.28) (0.61-0.93) (1.75-2.29) (3.38-4.16) (5.16-6.17) (7.38-8.67) (9.65-11.26) (12.21-14.28) (15.09-18.00)
0.13 0.33 1.65 4.07 7.48 11.96 17.02 22.69 28.81 36.15
≥75 15,380
(0.08-0.20) (0.25-0.44) (1.45-1.86) (3.75-4.41) (7.03-7.97) (11.35-12.60) (16.24-17.84) (21.71-23.71) (27.57-30.10) (34.38-37.98)
0.16 0.34 1.82 3.95 6.46 9.50 12.68 15.93 20.05 24.21
All 13,760
(0.11-0.24) (0.26-0.46) (1.60-2.06) (3.62-4.31) (6.02-6.94) (8.93-10.10) (11.97-13.43) (15.06-16.84) (18.93-21.23) (22.70-25.80)
0.84 1.62 2.27 3.33 4.49 5.07 7.14 7.14
<55 1,479 0 0
(0.48-1.47) (1.07-2.45) (1.58-3.27) (2.41-4.58) (3.34-6.03) (3.79-6.77) (5.24-9.70) (5.24-9.70)

0.15 0.31 1.12 2.33 3.66 5.00 6.01 7.44 9.91 11.72
55 to 64 2,613
Male
(0.06-0.41) (0.15-0.62) (0.77-1.61) (1.78-3.03) (2.93-4.56) (4.09-6.09) (4.96-7.27) (6.13-9.02) (8.08-12.14) (9.30-14.73)
0.04 0.19 1.31 3.07 4.73 6.85 9.49 12.36 15.25 19.42
65 to 74 5,241
(0.01-0.15) (0.10-0.36) (1.03-1.66) (2.61-3.60) (4.14-5.41) (6.08-7.71) (8.50-10.58) (11.12-13.73) (13.68-16.97) (17.16-21.94)
0.36 0.66 3.16 6.75 11.68 17.56 23.47 29.40 37.13 44.83
≥75 4,427
(0.22-0.59) (0.46-0.95) (2.68-3.73) (6.01-7.58) (10.65-12.80) (16.23-18.98) (21.84-25.21) (27.41-31.50) (34.59-39.81) (41.41-48.40)
Table 3.S20, Figure 3.S10 and Figure 3.S11 describe replacements is of concern as this now represents
the mortality of patients receiving a primary shoulder a significant proportion of all primary replacements.
replacement up to eight years following the primary The lack of completeness hampers one of the
procedure by gender and age group of the patients core functions of the registry, which is to provide a
undergoing surgery for elective indications only. comprehensive record of all implanted prostheses.
Data is shown at 30 and 90 days following the index
procedure in Table 3.S20 and then every year until the There are now 50,255 shoulder replacements eligible
eighth year. Mortality differences between the genders for analysis after the application of our data cleaning
are small and while males have higher mortality within processes. Patterns of use and the completeness of
the first five years following surgery, mortality in the data are becoming clearer and revision rates out to
longer term appears more comparable, see Figure eight years can be analysed. PROMs data continue
3.S10. When mortality is further divided by age (see to be collected so that patient outcomes in terms of
Figure 3.S11), it is clear that older males have higher pain and function can also be assessed alongside
mortality than females, this pattern first becomes revision rates. It has previously been identified that
evident after the age of 65. some patients who have worse post-operative PROMs
scores, i.e. a poor outcome, are not captured by the
3.6.5 Conclusions metric of revision surgery.
In this year’s report, we provide new and extensive Confirmed reverse polarity total shoulder replacement
insight into the use and performance of shoulder made up 60.2% of all shoulder replacements in 2019
constructs used in primary shoulder replacements and and the patterns of use observed in previous reports
also give a detailed description of revision rates by continue. This high level of use across indications
the indication for surgery. A detailed description of the indicates a growing confidence in this implant and
longitudinal PROMs data collection is also provided for a rapid change of practice in the NJR’s operational
both elective and trauma patients. geographical areas, despite limited high-level outcome
evidence. Proximal humeral hemiarthroplasties, and to
The pattern of use of primary shoulder replacements some extent conventional total shoulder replacements,
has continued to be documented. This year, we are declining in numbers.
have extensively revised shoulder implant data
processing and, building on the recent internal and Revision rates this year do not alter the pattern
external validation, it is now possible to report at the observed last year. Revision rates in younger patients
level of the construct. This detailed level of reporting continue to be high and are now 11.3% and 10.1%
has led to new and interesting insights, but it has in males and females respectively at five years.
also highlighted some inconsistencies within data These revision figures should be addressed in clinical
recorded in the registry, such as the unconfirmed discussions with younger patients wishing to undergo
procedures that are now reported. These are shoulder replacement surgery.
procedures where the reported patient procedure
At present, reverse polarity total shoulder replacement
disagrees with the implanted prostheses or there are
demonstrates the lowest revision rates at eight
insufficient elements recorded to verify a coherent joint
years. However, it is worth highlighting that these
replacement construct. The volume of unconfirmed
procedures have a higher early revision rate compared
proximal humeral hemiarthroplasty is consistently low,
to stemmed conventional total shoulder replacements,
and the volume of unconfirmed conventional total
until approximately two years following surgery.
shoulder replacements has fallen since the start of the
After two years the revision rate of stemmed reverse
registry. However, the volume of unconfirmed reverse
polarity shoulder replacements falls below stemmed
polarity total shoulder replacements is consistently
conventional total shoulder replacements. The
high and has increased in recent years. The volume
observed non-proportionality between conventional
of unconfirmed reverse polarity total shoulder
and reverse bearings combined with the differing strategies need to be developed nationally to improve
indications between the two procedures does not this low compliance. The post-operative PROMs
necessarily mean that reverse polarity shoulder are administered directly to patients on the NJR’s
replacements should be favoured over conventional behalf by their authorised contractor, NEC Software
total shoulder replacement, particularly for indications Solutions, and consideration of how many people
that would normally indicate the latter. respond and the timing of when they respond is now
also being addressed. The completeness of measures
More elective proximal humeral hemiarthroplasties cross-sectionally and importantly from a longitudinal
are being revised after the first year of surgery, with perspective and how this has changed across the
stemmed hemiarthroplasty seeming to outperform years has been described. A pre-operative and
either resurfacing or stemless hemiarthroplasty. 6-month matched elective cohort of 3,761 patients
While it may be argued that the higher revision rate is now available for analysis, but the representative
is mediated by the ease of the revision procedure, nature of this data compared to the whole cohort is
the PROMs data evidenced in this report does not not clear. It illustrates, in those who completed the
support this. The change in PROMs score between PROMs, that shoulder replacement surgery results
the pre-operative and 6-month assessment following in substantial improvement in both pain and function
surgery suggests less improvement and that the group for patients. However, it is less clear how those who
of patients that receive a hemiarthroplasty report less do not complete the PROMs fare, and the revision
positive outcome measures with the primary operation rate of those who do not respond to the PROMs
compared to others. questionnaires does appear to be different and higher,
when it is compared to those who do respond.
We suggest that more in-depth analysis which
accounts for case-mix should be conducted as, The largest gains by elective patients can be observed
while the age and gender distribution is similar, the in those patients receiving a conventional total
distribution of indications for which patients undergo shoulder replacement, followed closely by those
proximal humeral hemiarthroplasty is different to that receiving a reverse polarity shoulder replacement,
of either conventional total shoulder replacement or which is thereafter followed by those receiving a
reverse polarity shoulder replacement, with a much proximal humeral hemiarthroplasty.
higher proportion of patients indicating avascular
necrosis. An in-depth analysis accounting for the Overall, in this section of the report we have shown
variety of indications collected by the registry and that the volume of shoulder replacement surgery
other clinically relevant factors may help surgeons in the registry continues to grow rapidly and now
select different treatment modalities for patients. presents an opportunity for outcomes to be assessed
both by revision rates and by PROMs, although
This year we have presented a detailed description careful consideration of the latter in respect to its
of PROMs data with reference to not only those who generalisability is required. Importantly, our new
have responded, but the entire cohort of patients approach of whole construct validation using new
receiving a primary shoulder replacement. The pre- classifications and component attributes will lead
operative scores are administered and collected by to more meaningful analysis and provision of useful
hospital trusts and our analysis demonstrates that information for patients, surgeons and other
hospital trust compliance is poor. Better collection interested stakeholders.
3.7 In-depth
studies
3.7.1 The effect of surgical approach mortality and patient reported outcome measures
(PROMs) data with up to 13.75 years follow-up were
in total hip replacement on outcomes: analysed. There were seven surgical approach groups:
an analysis of 723,904 elective conventional posterior, lateral, anterior and trans-
operations from the National Joint trochanteric groups and minimally invasive posterior,
Registry for England, Wales, Northern lateral and anterior. Operations with metal-on-metal
bearings were excluded from analysis.
Ireland and the Isle of Man
Survival methods were used to compare revision
Ashley W Blom, Linda P Hunt, Gulraj S Matharu,
rates and 90-day mortality. Groups were compared
Michael R Reed, Michael R Whitehouse
using Cox proportional hazards and Flexible
BMC Med 18, 242 (2020) DOI: https://doi. Parametric Survival Modelling (FPM). Confounders
org/10.1186/s12916-020-01672-0 included age at surgery, sex, risk group (indications
additional to osteoarthritis), ASA grade, THR fixation,
Reproduced in summary form under CC BY thromboprophylaxis, anaesthetic, body mass
4.0 licence. index and deprivation. PROMs were analysed with
regression modelling or non-parametric methods.
Background
Total hip replacement (THR) is both clinically and Findings
cost-effective. The surgical approach that is employed A total of 12,989 (1.8%) of 723,904 implants were
influences the outcome, however there is little revised during follow-up; 84,294 (11.6%) died without
generalisable and robust evidence to guide practice. undergoing revision. Figure 3.1 shows the estimated
cumulative percentage revised (Kaplan-Meier) up to 12
Methods years for the seven groups.
A total of 723,904 primary THRs captured in the
National Joint Registry, linked to hospital inpatient,
Figure 3.1 Cumulative percentage revised (Kaplan-Meier) up to 12 years for the seven surgical
approach groups.
Cumulative % probability of revision
0
0 1 2 3 4 5 6 7 8 9 10 11 12
Years from primary
No. at risk
Ant/Other no MI 26582 11686 522
Ant/Other w MI 3830 1093 82
Lat/Ant−lat/H no MI 260415 134521 31312
Lat/Ant−lat/H w MI 9101 4995 1358
Post no MI 394735 143390 20401
Post w MI 22663 9443 1840
Trans−troch no MI 6578 4868 1975
Ant/Other no MI Ant/Other w MI
Lat/Ant−lat/H no MI Lat/Ant−lat/H w MI
Post no MI Post w MI
Trans−troch no MI
Unadjusted analysis showed a higher revision risk than

the referent conventional posterior for the conventional
lateral, minimally invasive lateral, minimally invasive
anterior and trans-trochanteric groups. This persisted
with all adjusted FPM and adjusted Cox models
(see Table 3.1 overleaf), except in the Cox model
including BMI where the higher revision rate persisted
for the conventional lateral approach (HRR 1.12
[95% CI 1.06,1.17] P<0.001). PROMs demonstrated
statistically, but not clinically, significant differences.
Self-reported complications were more frequent with
the conventional lateral approach.
Table 3.1 Regression models to compare approach groups for revision risk (n=723,747 with complete information).
(i) Cox proportional hazards regression models, with stratification by age/sex/risk groups.
With adjustment
for fixation, ASA
With adjustment for and BMI sub-group
Minimally invasive HRR fixation and ASA (n=443,657)
Approach procedure used [95% CI] HRR [95% CI] HRR [95% CI]
Posterior No 1 [Referent] 1 [Referent] 1 [Referent]
Posterior Yes 0.99 [0.89,1.10] 0.92 [0.83,1.02] 0.89 [0.77,1.02]
P=0.864 P=0.110 P=0.097
Lat/Ant-Lat/Hard No 1.05 [1.01,1.09] 1.07 [1.03,1.11] 1.12 [1.06,1.17]
P=0.009 P=0.001 P<0.001
Lat/Ant-Lat/Hard Yes 1.31 [1.16,1.50] 1.28 [1.13,1.46] 1.02 [0.80,1.30]
P<0.001 P<0.001 P=0.861
Ant/Other No 1.04 [0.95,1.14] 1.03 [0.94,1.13] 1.01 [0.88,1.15]
P=0.431 P=0.561 P=0.921
Ant/Other Yes 1.67 [1.36,2.05] 1.48 [1.21,1.82] 1.03 [0.71,1.51]
P<0.001 P<0.001 P=0.870
Trans-trochanteric No 1.22 [1.07,1.40] 1.40 [1.22,1.60] 1.48 [1.14,1.91]
P=0.004 P<0.001 P=0.003
Additional pairwise comparisons:
Lat/Ant-Lat/Hard No vs. Yes P=0.001 P=0.005 P=0.475
Ant/Other No vs. Yes P<0.001 P=0.001 P=0.902
Note: Lat/Ant-Lat/Hard = Lateral / Anterolateral / Hardinge. Ant/Other = Anterior / Other. ASA = American Society of Anesthesiologists Physical Status. BMI = body
mass index.
(ii) FPM models, with adjustment for time-varying effects of age, sex, risk group.
With adjustment for fixation and
Minimally invasive ASA, as time-varying effects
Approach procedure used Coefficent [95% CI] Coefficient [95% CI]
Posterior No 0 [Referent] 0 [Referent]
Posterior Yes -0.006 [-0.109,0.096] P=0.903 -0.081 [-0.183,0.022] P=0.125
Lat/Ant-Lat/Hard No 0.056 [0.019,0.093] P=0.003 0.069 [0.031,0.106] P<0.001
Lat/Ant-Lat/Hard Yes 0.282 [0.154,0.411] P<0.001 0.264 [0.135,0.392] P<0.001
Ant/Other No 0.031 [-0.063,0.126] P=0.516 0.019 [-0.075,0.114] P=0.688
Ant/Other Yes 0.516 [0.311,0.721] P<0.001 0.380 [0.174,0.585] P<0.001
Trans-trochanteric No 0.213 [0.075,0.350] P=0.002 0.309 [0.170,0.448] P<0.001
Lat/Ant-Lat/Hard No vs. Yes P=0.001 P=0.003
Ant/Other No vs. Yes P<0.001 P=0.002
Note: Lat/Ant-Lat/Hard = Lateral / Anterolateral / Hardinge. Ant/Other = Anterior / Other. ASA = American Society of Anesthesiologists Physical Status.
Our previous work on mortality after hip replacement mortality than the conventional posterior approach
had identified confounding factors and a series of (HRR 1.15 [95% CI 1.01-1.30] P=0.029 in the fully
univariable analyses confirmed these. Thus, in our adjusted model). There were no other significant
analysis shown in Table 3.2, we have adjusted for differences in mortality compared to the referent
these factors. In all models, the conventional lateral conventional posterior approach group.
approach was associated with a higher risk of
Table 3.2 Cox ‘proportional hazards’ regression model to compare 90-day mortality between the seven approach sub-groups.
(vi) With covariate
(v) With covariate adjustment for
(iv) With covariate adjustment for sex, age, ASA,
adjustment for sex, age, ASA, year of primary,
sex, age, ASA, year of primary, fixation, mechanical
year of primary, fixation, mechanical and chemical
fixation, mechanical and chemical thromboprophylaxis,
and chemical thromboprophylaxis, anaesthetic group,
thromboprophylaxis anaesthetic group quintile of area
(ii) With covariate (iii) With covariate and anaesthetic and quintile of deprivation and
adjustment for adjustment for sex, group, stratified area deprivation, BMI subgroup,
sex, age, ASA and age, ASA, year of by ‘risk group’ stratified by ‘risk stratified by ‘risk
year of primary, primary and fixation, (n=713,994 group’ (n=572,719 group’ (n=359,883
Minimally (i) Unadjusted stratified by ‘risk stratified by ‘risk with complete with complete with complete
invasive model (n=723,747; group’ (n=723,747; group’ (n=723,747; information; 2,621 information; 2,192 information; 1,148
procedure Number for 2,673 deaths) 2,673 deaths) 2,673 deaths) deaths) deaths) deaths)
Approach used analysis HRR [95% CI] HRR [95% CI] HRR [95% CI] HRR [95% CI] HRR [95% CI] HRR [95% CI]
Posterior No 394,655 1 [referent] 1 [referent] 1 [referent] 1 [referent] 1 [referent] 1 [referent]
Posterior Yes 22,655 0.80 [0.62-1.04] 0.88 [0.68-1.14] 0.90 [0.69-1.17] 0.91 [0.70-1.18] 0.93 [0.68-1.27] 0.87 [0.57-1.34]
P=0.101 P=0.344 P=0.447 P=0.475 P=0.644 P=0.525
Lat/Ant-Lat/Hard No 260,353 1.38 [1.27-1.49] 1.14 [1.05-1.24] 1.14 [1.05-1.23] 1.16 [1.06-1.26] 1.11 [1.01-1.21] 1.15 [1.01-1.30]
P<0.001 P=0.002 P=0.002 P=0.001 P=0.027 P=0.029
Lat/Ant-Lat/Hard Yes 9,099 1.21 [0.87-1.69] 1.02 [0.73-1.43] 1.03 [0.74-1.44] 0.91 [0.63-1.31] 0.87 [0.58-1.31] 0.80 [0.41-1.55]
P=0.252 P=0.896 P=0.847 P=0.605 P=0.512 P=0.507
Ant/Other No 26,578 1.03 [0.83-1.28] 0.96 [0.77-1.19] 0.96 [0.77-1.19] 0.96 [0.78-1.20] 0.94 [0.75-1.19] 0.89 [0.63-1.26]
P=0.793 P=0.703 P=0.715 P=0.732 P=0.620 P=0.509
Ant/Other Yes 3,829 0.80 [0.43-1.50] 0.88 [0.47-1.65] 0.91 [0.49-1.71] 0.92 [0.49-1.71] 1.23 [0.61-2.47] 1.01 [0.32-3.15]
P=0.493 P=0.697 P=0.780 P=0.787 P=0.563 P=0.986
Trans-trochanteric No 6,578 1.49 [1.05-2.11] 1.19 [0.83-1.69] 1.16 [0.81-1.65] 1.16 [0.81-1.67] 1.12 [0.75-1.68] 1.12 [0.52-2.41]
P=0.026 P=0.342 P=0.412 P=0.409 P=0.571 P=0.771
Lat/Ant-Lat/Hard No vs. Yes P=0.452 P=0.517 P=0.573 P=0.195 P=0.250 P=0.283
Ant/Other No vs. Yes P=0.460 P=0.807 P=0.884 P=0.885 P=0.447 P=0.833
Note: Lat/Ant-Lat/Hard = Lateral / Anterolateral / Hardinge. Ant/Other = Anterior / Other. ASA = American Society of Anesthesiologists Physical Status. BMI = body mass index.
309
Interpretation
Lateral approaches for THR are associated with worse
outcomes, including more deaths and revisions,
than the posterior approach. The conventional
lateral approach (36.0%) is the second most popular
approach and is currently used annually in over 20,000
primary THRs in the registry. This approach was
associated with worse outcomes in all measures than
the commonest approach, the conventional posterior
(54.5%). The data presented here does not support its
continued use over alternatives.
It would be difficult and perhaps unwise to attempt

conversion of experienced surgeons to an approach
with which they may be unfamiliar. However,
surgeons in training should be taught alternative
approaches to the lateral associated with better
outcomes. The data does support continued use
of minimally invasive approaches, with acceptable
mortality and PROMs outcomes, although minimally
invasive lateral and anterior approaches may be
associated with higher revision rates than their
corresponding conventional approaches.
3.7.2 What are the inpatient and day oral antibiotics. Our aim was to estimate the cost to
the English NHS of inpatient and day case admissions,
case costs following primary total hip in the five years following primary THR, of patients
replacement of patients treated for who were treated with a one- or two-stage revision
prosthetic joint infection: a matched for PJI following primary THR (revised PJI patients
cohort study using linked data from hereinafter), compared to patients whose THR was
either not revised or revised for reasons not related to
the National Joint Registry and
PJI (comparator patients hereinafter).
Hospital Episode Statistics
Methods
Kirsty Garfield, Sian Noble, Erik Lenguerrand,
Michael R. Whitehouse, Adrian Sayers, Mike R. Reed, This matched cohort study utilised linked NJR and
Ashley W. Blom Hospital Episodes Statistics (HES) inpatient and
day case admission data, from 1 April 2003 until
BMC Medicine 2020; 18:335. DOI: https://doi. 1 December 2014. HES data includes admissions
org/10.1186/s12916-020-01803-7 in England funded by the English NHS, as such
the analysis was limited to patients receiving NHS
Reproduced in summary form under CC BY funded treatment in England. Eligibility criteria for
4.0 licence. inclusion in the revised PJI group included: infection
as an indication for revision recorded in the NJR;
This research was funded by the National Institute for
a one-stage revision or at least part one of a two-
Health Research (NIHR) under its Programme Grants
stage revision for PJI, between 2006 and 2009, that
for Applied Research programme (RP-PG-1210-
occurred within five years of the primary THR; first
12005). The funding body had no role in the design of
revision for PJI on the index side; no revision PJI
the study and collection, analysis and interpretation of
surgery on the non-index side during the follow-up
data and in the writing of the manuscript. This study
period and complete matching variables. Eligibility
was supported by the NIHR Biomedical Research
criteria for inclusion in the comparator group included:
Centre at University Hospitals Bristol NHS Foundation
a primary THR between the dates of the primary THRs
Trust and the University of Bristol. The views
of revised PJI patients; no revision for PJI on the index
expressed in this publication are those of the author(s)
side reported in the NJR; no revision surgery for PJI
and not necessarily those of the NHS, the National
on the non-index side during the follow-up period and
Institute for Health Research or the Department of
complete matching variables.
Health and Social Care.
Patients were matched using a combination of exact
Background
and radius (close) matching, using a one to five
Following total hip replacement (THR), a small ratio of revised PJI to comparator patients. Patient
percentage of patients develop a periprosthetic characteristics and primary THR surgery factors that
joint infection (PJI). PJI is a serious and debilitating were considered to potentially impact the likelihood of
complication which is associated with a negative PJI following THR were included. Exact matching was
impact on morbidity and quality of life and an used for sex, ASA grade, type of replacement (total or
increased risk of mortality. Compared to primary THR resurfacing) and hospital. Radius matching was used
and aseptic revision, revision procedures for PJI are for date of primary THR (one year radius) and age (ten
associated with an increased burden on healthcare year radius).
providers due to longer operating times, higher
readmission rates, costly repeat procedures, extended All inpatient and day case admissions (not limited
hospital admissions, more hospital outpatient to orthopaedic admissions) reported in HES during
appointments, and prolonged use of intravenous and the five years following primary were included. HES
data was cleaned and costs were estimated for each
admission using Healthcare Resource Groups and Results
corresponding NHS reference costs, which are based
Between 2006 and 2009, 1,914 one- or two-stage
on average unit costs of NHS providers.
revisions for PJI were identified in the NJR. From
A two-part model, which accounted for clustering these, 422 patients met the inclusion criteria, had
of revised PJI and comparator patients within their NJR data that could be linked to HES data and were
matching group and excess zeros, was used to matched to 1,923 comparator patients. There was
estimate the difference in number of stays and costs balance between revised and comparator patients for
between revised PJI and comparator patients. Age, variables where exact matching was employed (see
sex, ASA grade, diagnosis of osteoarthritis, operation Table 3.3). Other variables were moderately balanced
date, Charlson Comorbidity Index, bearing surface and between the two groups and were subsequently
procedure were controlled for within the model. adjusted for in the analysis model.
Table 3.3 Characteristics of matched patients revised and not revised for PJI following primary THR.
Revised PJI group Comparator group
(n=422) (n=1,923)
Characteristics Number (%) Number (%)
Date of primary – range 16/05/03 - 02/12/09 28/04/03 - 01/12/09
Age - mean (range) 66 (21-95) 67 (23-92)
Female 191 (45) 891 (46)
Osteoarthritis diagnosis 398 (94) 1,862 (97)
ASA grade
P1 69 (16) 302 (16)
P2 298 (71) 1,399 (73)
P3 55 (13) 222 (12)
Charlson
0 275 (65) 1,415 (74)
1 97 (23) 333 (17)
2 31 (7) 104 (5)
3 or above 19 (5) 71 (4)
Procedure
Cemented 164 (39) 782 (41)
Uncemented 158 (37) 668 (35)
Hybrid/Reverse hybrid 64 (15) 324 (17)
Resurfacing 36 (9) 149 (8)
Bearing type
Metal-on-plastic 254 (60) 1,179 (61)
Metal-on-metal 99 (23) 354 (18)
Ceramic-on-ceramic 41 (10) 199 (10)
Ceramic-on-plastic/ metal-on-ceramic/ ceramic-on-metal 28 (7) 191 (10)
Matches per revised PJI patient
5 matching comparator patients 358 (85)
During the five years following primary THR, revised primary THR was £41,633 (95% CI £39,079 to
PJI patients had eight admissions on average, £44,187) for revised PJI patients and £8,181 (95% CI
compared to an average of three admissions for £7614 to £8748) for comparator patients, equating to
comparator patients. The average cost of inpatient a difference in costs of £33,452 (95% CI £30,828 to
and day case admissions in the five years following 36,077; p < 0.00) (see Table 3.4).
Table 3.4 Average total and annual inpatient and day case hospital admission costs over the five years following
THR, by revised PJI and comparator patients.
Revised PJI group (n=422) Comparator group (n=1,923)
Adjusted cost Adjusted cost Adjusted difference in costs
Years £ Mean (SE) £ Mean (SE) £ (95% CI)
1st year post-primary 14,686 (816) 1,959 (111) 12,727 (11,094 to 14,360)
2nd year post-primary 10,575 (682) 1,503 (91) 9,071 (7,719 to 10,424)
3rd year post-primary 6,974 (580) 1,512 (97) 5,462 (4,306 to 6,618)
4th year post-primary 5,168 (501) 1,584 (131) 3,584 (2,611 to 4,557)
5th year post-primary 4,427 (431) 1,568 (101) 2,859 (1,999 to 3,720)
Total over five years 41,633 (1,303) 8,181 (289) 33,452 (30,828 to 36,077)
Note: Marginal means after adjusting for excess zero; adjusted for age, sex, ASA grade, diagnosis of osteoarthritis, operation date, Charlson Comorbidity Index,
bearing surface and procedure.
Note: SE = Standard Error.
Discussion identified. The richness of the NJR dataset meant that

most known confounders were included as matching
Over the five years following primary THR, patients
variables or controlled for within the regression. The
who had a one- or two-stage revision THR for PJI had
exception was body mass index, which was not
more hospital admissions than comparator patients on
included due in part to it not being included in earlier
average, at an estimated additional cost of £33,452.
NJR data collection forms.
Relative to other studies exploring the costs of PJI
treatment, the sample size was large. Inpatient and day case admissions for any indication
were included, as it was acknowledged that PJI
The inclusion criteria meant that a minority of
may affect other areas of patients’ lives, leading to
comparator patients may have developed a PJI and
admissions for reasons not directly related to the PJI.
received alternative treatments. The estimated cost
Including outpatient, primary and community care,
burden therefore does not compare infected with
prescribed medications and admissions funded by
uninfected patients but compares those revised for
the NHS outside of England would result in increased
PJI with a one- or two-stage revision compared to
estimates of the financial burden of treating PJI.
those not revised for PJI. As the indication for revision
is defined at the time of revision, revisions attributed Conclusion
to infection could be an under- or overestimate, as
intraoperative results may alter the opinion of the This study showed that patients who develop PJI and
treating surgeon. have revision surgery cost approximately £33,000
(over five-fold) more than patients not revised for PJI,
We were able to match 94% of revised PJI patients based on their hospital admissions alone. As demand
to comparator patients. Most revision PJI patients for primary and revision THR is predicted to rise in
were matched to five comparator patients; however, future, future research should focus on finding ways to
to maximise the sample size, revision PJI patients reduce the incidence of PJI following THR and finding
were still included if less than five matches were cost-effective treatments for PJI.
3.7.3 Effect of Bearing Surface on Polyethylene irradiated with an irradiation dose of five
or more Mrad was classified as crosslinked (XLPE).
Survival of Cementless and Hybrid The bearing combinations analysed were: ceramic
Total Hip Arthroplasty: Study of Data on polyethylene (CoP), metal on polyethylene (MoP),
in the National Joint Registry for ceramic on cross-linked polyethylene (CoXLPE), metal
England, Wales, Northern Ireland and on cross-linked polyethylene (MoXLPE), ceramicized
metal on cross-linked polyethylene (CMoXLPE), and
the Isle of Man
ceramic on ceramic (CoC). The primary endpoint
Davis, Edward T; Pagkalos, Joseph; Kopjar, Branko. was revision for any reason. Additional analyses were
performed to investigate specific reasons for revision,
JBJS Open Access 2020 May 15;5(2):e0075 DOI: such as infection, aseptic loosening, wear, dislocation,
https://doi.org/10.2106/JBJS.OA.19.00075 periprosthetic fracture, pain and implant fracture.
Reproduced in summary form under CC BY-NC-ND Statistical analysis

4.0 licence.
Kaplan-Meier analysis adjusted for the competing
Background risk of death was used for overall and cause-specific
revisions. Revisions for other reasons were treated as
The effect of the bearing surface on total hip a competing risk in cause-specific analyses. A Cox
replacement (THR) survival has received a lot of proportional hazards regression model was used to
attention over the last two decades and features obtain hazard ratios accounting for a competing risk
in every report of the National Joint Registry (NJR) of death.
and other registries around the world. Polyethylene-
based bearings have traditionally been associated Results
with particle-related osteolysis which led to the
Overall risk of revision
development and more widespread use of hard on
hard bearings. The development of cross-linked When all patients were analysed (adjusted for a
polyethylene (XLPE) has led to a marked reduction competing risk of death), the lowest cumulative
in the risk of revision of THRs utilising this bearing. In incidence of revision for any reason at ten years of
the annual NJR reports, the type of polyethylene is follow-up was 1.96% for CMoXLPE (95% CI 1.35-
not stratified which leads to XLPE being merged with 2.76), followed by 2.52% (95% CI 2.14-2.95) for
conventional PE and presented as a single group. CoXLPE, 2.81% (95% CI 2.58-3.05) for MoXLPE,
3.03% (95% CI 2.75-3.33) for CoP, 3.47% (95% CI
Methods 3.29-3.65) for CoC, and 3.53% (95% CI 3.37-3.70) for
The NJR dataset and data on polyethylene MoP (see Figure 3.2).
manufacturing characteristics were used for the
study. Primary THRs implanted between 1 January
2004 and 28 July 2016 were eligible for analysis.
Figure 3.2 Cumulative incidence of revision for any reason by bearing combination (P<0.0001).
Patients under the age of 55

When patients under the age of 55 at the time of
implantation were analysed independently, the
cumulative incidence of revision at ten years was
1.80% (95% CI 1.11-2.78) for CMoXLPE, 3.16% (95%
CI 2.36-4.13) for CoP, 3.35% (95% CI 2.16-4.95) for
CoXLPE, 4.34% (95% CI 3.95-4.76) for CoC, 5.20%
(95% CI 3.11-8.05) for MoXLPE, and 6.12% (95% CI
4.97-7.42) for MoP (see Figure 3.3 overleaf).
Figure 3.3 Cumulative incidence of revision for any reason by bearing combination in patients aged
less than 55 years of age at the time of primary THR (P<0.0001).
Reasons for revision Multivariate analysis

THRs with CMoXLPE and CoXLPE bearings The Cox regression model revealed CMoXLPE and
demonstrated the lowest risk of revision due to aseptic CoXLPE as the bearings with the biggest reduction
loosening. When revision for infection was analysed, in the risk of revision (see Table 3.5). A similar trend
CoP and CoC THRs had a lower risk of revision when was observed when patients under the age of 55 were
compared to MoP THRs as the reference. A total of analysed independently. An additional Cox model
1.3/1000 implants with a ceramic bearing were revised was built to include head size. Due to low numbers
for a ceramic liner fracture. of THRs with large heads and conventional PE, the
bearings analysed were limited to CMoXLPE, CoXLPE,
MoXLPE and CoC (see Table 3.6 on page 318). The
lowest risk of revision for any reason was seen in
THRs with CMoXLPE and CoXLPE in the model.
Table 3.5 Cox regression hazard ratios of risk of any revision by bearing combination.
Characteristic/Level All ages (HR) <55 years (HR)

Age
55 to <64 years 0.85 (0.79-0.91)
65 years to <75 years 0.73 (0.68-0.79)
75 years and more 0.68 (0.62-0.73)
<55 years 1.0 (reference)
Gender
Male 1.18 (1.13-1.23) 1.20 (1.08-1.34)
Female 1.0 (reference) 1.0 (reference)
Bearing combination
CoC 0.77 (0.72-0.82) 0.64 (0.52-0.78)
CoP 0.74 (0.66-0.82) 0.50 (0.36-0.70)
CoXLPE 0.66 (0.60-0.72) 0.61 (0.47-0.78)
MoXLPE 0.81 (0.76-0.87) 0.77 (0.59-1.01)
CMoXLPE 0.58 (0.48-0.71) 0.47 (0.30-0.76)
MoP 1.0 (reference) 1.0 (reference)
Stem fixation
Cementless 1.35 (1.28-1.42) 1.45 (1.26-1.68)
Cemented 1.0 (reference) 1.0 (reference)
Table 3.6 Cox regression hazard ratios of risk of any revision by bearing combination and head size.
Characteristic/Level All causes (HR) Dislocation (HR) Aseptic loosening (HR)

Age
55 to <64 years 0.82 (0.76-0.89) 0.84 (0.70-1.02) 0.79 (0.68-0.91)
65 years to <75 years 0.76 (0.70-0.82) 0.86 (0.71-1.04) 0.61 (0.52-0.71)
75 years and older 0.76 (0.69-0.84) 1.01 (0.82-1.26) 0.43 (0.35-0.53)
<55 years 1.0 (reference) 1.0 (reference) 1.0 (reference)
Gender
Male 1.16 (1.10-1.23) 1.05 (0.93-1.19) 1.34 (1.19-1.50)
Female 1.0 (reference) 1.0 (reference) 1.0 (reference)
Bearing combination
CoC 0.99 (0.93-1.07) 0.84 (0.72-0.99) 1.05 (0.90-1.21)
CoXLPE 0.84 (0.77-0.92) 0.90 (0.61-1.34) 0.85 (0.70-1.03)
CMoXLPE 0.75 (0.62-0.92) 0.90 (0.61-1.34) 0.52 (0.32-0.860
MoXLPE 1.0 (reference) 1.0 (reference) 1.0 (reference)
Stem fixation
Cementless 1.33 (1.25-1.42) 1.03 (0.91-1.18) 2.26 (1.93-2.65)
Cemented 1.0 (reference) 1.0 (reference) 1.0 (reference)
Head size
≤28 mm 1.07 (0.99-1.15) 2.13 (1.82-2.48) 0.85 (0.73-1.00)
32 mm 0.92 (0.86-0.98) 1.27 (1.09-1.47) 0.79 (0.69-0.91)
≥36 mm 1.0 (reference) 1.0 (reference) 1.0 (reference)
Discussion Conclusion
Our analysis revealed that XLPE bearing THRs had XLPE-based bearing THRs were associated with a
a significantly lower risk of revision when compared marked reduction in the risk of revision at a maximum
to conventional PE (MoP). Ceramicized metal on follow-up of 13 years.
XLPE and ceramic on XLPE were associated with the
lowest risk of revision for any reason in our multivariate
analysis. Due to the marked difference in the risk
of revision between conventional polyethylene and
crosslinked polyethylene we recommend stratification
of the polyethylene-based bearings when comparing
survival of THRs.
3.7.4 Provision of revision knee specialist orthopaedic work. In response to GIRFT

recommendations, the British Association for Surgery
surgery and calculation of the effect of the Knee (BASK) set up a working group in
of a network service reconfiguration: revision KR surgery. Part of the group’s remit was to
An analysis from the National Joint undertake an exploratory analysis using summary NJR
Registry for England, Wales, Northern data to 1) describe the current provision of revision KR
in England and Wales at the individual surgeon and
Ireland and the Isle of Man
unit level; and 2) investigate the effect on workload of
N.S. Kalson; J.A. Mathews; J. Miles; B.V. Bloch; A.J. case distribution in a network model.
Price; J.R.A. Phillips; A.D. Toms; P.N. Baker; British
Association for Surgery of the Knee, Revision Knee Methods
Working Group A data extract was obtained from the National Joint
Registry for England, Wales, Northern Ireland and the
The Knee. Elsevier; 2020 Oct 1;27(5):1593–600. Isle of Man containing all knee procedures coded on
DOI: https://doi.org/10.1016/j.knee.2020.07.094 a K2 form during 2016, 2017 and 2018 (three years
total data). To calculate individual surgeon workload,
Reproduced in summary form with permission from
procedures performed at both NHS and independent
Elsevier B.V. and the authors.
sites were combined.
Background
The effect of service re-organisation into a network
Revision knee replacement (KR) is expensive, model was undertaken by assigning each NHS site
technically challenging and patients are at risk of to one of 13 geographical regions, corresponding
significant complications and poor outcomes. It is to regions used by the NJR. Based on their current
well-established in primary knee surgery (total and annual revision workload, units within each region
partial) that higher surgeon volume is associated with were categorised as either a Major Revision Centre
lower complication rates, lower revision rates and (MRC), Revision Unit (RU) or Primary Arthroplasty Unit
lower mortality rates. Evidence is becoming available (PAU). MRCs were defined as those undertaking >210
for revision surgery; analysis of >17,000 revision hip revisions over three years (>70 revision KR per year).
cases showed that low volume centres (<13 cases Revision Units were defined as units undertaking an
per year) had significantly worse 90-day mortality and average of ≥70 procedures over three years, giving
1-year re-revision rates than high-volume centres an average minimum volume of >20 per year. These
and analysis of ~25,000 revision knee cases showed thresholds were set based on 1) analysis of literature
decreased re-revision rates in high volume centres. examining the relationship between volume and
Analysis of >30,000 hip and knee revision cases outcome that suggests units undertaking <25 KR per
showed lower complication rates and lower 90-day year have increased early complication rates, higher
re-admission rates in high volume centres. Although 90-day readmission rates and higher re-revision rates;
these reports point towards a volume-outcome and 2) a need to have at least one MRC and several
relationship in revision KR, the precise level remains a RUs in each region. For this analysis the threshold
research question. was reduced to 20 per unit to allow an increase in
unit volume after workload re-allocation to boost units
Centralisation of complex services has occurred above 25 cases per year. Primary Arthroplasty Units
within and beyond orthopaedics. Major trauma were all other units (<20 revision KR/year).
care has been organised into specialist trauma
units and major trauma centres, and in England the In total 25 MRCs, 82 RUs and 125 PAUs were
‘Getting It Right First Time’ (GIRFT) initiative has identified and used for calculations. Work currently
delivered recommendations on service organisation done by PAUs was re-assigned to MRCs and RUs
and infrastructure for units undertaking complex, evenly (number of units in a region divided by re-
allocated cases). For example, 100 cases at PAUs in a The median annual surgeon volume was 2.3±0.2
region with ten revision centres (two MRCs and eight cases per year (Range 0-56) (see Figure 3.4). The
RUs) would result in 10 additional cases per MRC/RU majority of surgeons performed small numbers of
per year. revision procedures with 1,020 (75%) surgeons
performing <7 revisions per year (see Table 3.7).
Main findings Overall, 209 of 1,353 surgeons performed ≥10/year
There were 20,857 revision KR procedures recorded (56% of total work, 15.8% of surgeons) and 100
on the NJR between 2016 and 2018 (three years) surgeons performed ≥15/year (36% of total work, 7%
across 232 NHS (18,355 cases, 88%) and 167 of surgeons). A total of 64 surgeons performed ≥20/
independent healthcare provider sites (2,502 cases, year and 19 performed ≥30/year. Overall, the highest
12%). In total 1,353 surgeons performed at least one volume 397 (29%) surgeons performed 75% of the
revision KR procedure over this time period. revision KR workload.
Figure 3.4 Individual surgeon volume.

Surgeons’ individual volume, including procedures undertaken at NHS and independent sites, over
three years. More than 1,300 surgeons undertook at least one revision KR procedure; 100 surgeons
carried out >44 procedures over three years, accounting for 37% of national procedure volume
(indicated by the black arrow).
Number of revisions performed over 3 years
Surgeon (ranked by total volume)
Table 3.7 Revision surgeon volume, average annual number of revisions (2016-2018).
Annual volume Number of surgeons

0–4 872
5–9 257
10 – 14 113
15 – 19 41
20 – 24 29
25 – 29 16
30 – 39 18
40+ 7
Total 1,353
The median NHS site annual volume over three years volume for each site (see Figure 3.5) shows that most
was 51 cases, equating to a mean of 18.3±1.3 cases centres had a large number of surgeons undertaking a
(range 1-211) per year. Overall 15 sites performed small number of procedures. Five sites were identified
≥70/year, 92 performed 20-69 and 125 performed (of 232, 2%) where mean surgeon volume across all
<20/year. Comparing surgeon number against revision revision surgeons was >10 cases per year.
Figure 3.5 Site versus surgeon volume.

The number of surgeons undertaking revision KR procedures at each site (NHS only) versus total site
volume (three years total data). The black line plots a threshold at an average of ten procedures for
each surgeon per year; 5/232 sites hit this threshold and sit above the line.
Number of revisions per site (total three year volume)
Number of Surgeons performing Revisions at the Site (total)
Even redistribution of caseload from PAUs to MRCs/ the range was 19-174, 18%-36%). Amongst 25 MRCs
RUs is shown in Table 3.8. Ten mid-volume units were and 85 RUs there was an average increase in MRC
made MRCs to allow each region to have at least one workload of 11 procedures (range 6-14) per year. All
MRC unit for the purpose of the model, giving 25 MRC MRCs and RUs had a revision rate for their primary
units for calculations. In total 1,235 revisions (21%) knee arthroplasties ‘as expected’ or ‘better than
were reallocated from PAUs (for each individual region expected’ (2014-2019).
Table 3.8 The effect of MRC-RU reconfiguration on centre volume.

Percentage No. of Average
Total of region revisions No. of Major No. of additional
revisions in Primary to be Revision No. of Primary revisions
in region Arthroplasty absorbed Centres Revision Arthroplasty per unit
Region (annual) Units (annual) (MRC) Units (RU) Units (PAU) (annual)
East Midlands 445 22% 80 1 5 7 13
Eastern 484 34% 124 2 8 8 12
London 627 29% 141 3 10 21 11
Mid and West Wales 88 27% 19 1 2 3 6
North East 397 21% 70 2 7 7 8
North West 654 36% 174 3 13 21 11
South Central 610 26% 126 2 7 8 14
South East Coast 500 25% 98 2 8 10 10
South East Wales 200 22% 36 1 2 4 12
South West 611 18% 94 2 8 7 9
West Midlands 746 20% 124 3 7 12 12
Yorkshire 594 33% 148 3 8 14 13
Total 5,956 21% 1,235 25 85 122 11
Discussion year) was, on average, 11 cases for MRC/RUs. An

additional effect of this change was that all revision
Revision KR surgery is expensive for healthcare
units would reach a volume threshold of 30 cases per
providers and challenging for the clinical team. It is well
year. Although this would appear to be a manageable
established that increasing surgeon and centre volume
additional workload, representing an approximate 20%
improves cost-effectiveness and outcomes. Here we
increase in revision KR procedures across the system
report the current provision of revision KR in England,
for centres that continue to perform knee revisions,
Wales and Northern Ireland using descriptive data
the precise redistribution of cases between MRCs and
from the NJR and describe the effect of implementing
RUs will depend on case complexity.
GIRFT ideals and moving revision KR from low volume
centres to higher volume units. There are limitations to the study. The NJR does
not allow stratification by case complexity; we are
We found many surgeons performing a small number
unable to distinguish between the conversion of
of procedures; more than 1,000 surgeons performed
a unicompartmental replacement to a total knee
<7 procedures per year. In addition to a large number
replacement, from a complex revision for infection
of low volume surgeons there were a large number
requiring stems, augments and extensor mechanism
of low volume units. This finding is similar to work
reconstruction. We therefore do not understand the
published almost ten years ago describing volumes
case-mix for these low volume surgeons and units.
of revision knee procedures using NJR data and
demonstrates little has changed over the last decade Conclusions
despite the introduction of clinical networks and the
GIRFT initiative. Data presented here demonstrates that currently in
England and Wales a number of surgeons undertake
To help further the discussion around service a small number of revision KR procedures in a
reconfiguration we developed a hypothetical model large number of low volume units. High volume
using thresholds for unit volume to investigate the surgeons and centres do already exist, creating a
effect of a network model for revision KR work. This pre-fabricated network for the implementation of a
had the aim of minimising the number of procedures network model of care. Creating referral centres from
performed in low volume centres by low volume low volume units and redistributing this work showed
surgeons and was achieved by distributing these that the scale of the uplift would be manageable
complex cases across the regional network. The and would have the positive effect of raising centres
estimated alteration in work resulting from creation of above a 30/year threshold.
122 PAUs (with current volumes <20-30 cases per
3.7.5 The association between we investigated the association between consultant
surgical volume in the year (365 days) prior to the index
surgical volume and failure of primary operation of interest, and the risk of revision in patients
total hip replacement in England and undergoing elective primary THA between 1 April 2003
Wales: Findings from a prospective and 22 February 2017.
national joint replacement register The primary outcome of interest was all cause revision
Adrian Sayers, Fiona Steele, Michael R Whitehouse, after a primary THA.
Andrew Price, Yoav Ben-Shlomo, Ashley W Blom
The primary exposure of interest in this study was
BMJ Open 2020;10:e033045 DOI: https://doi. the consultant surgical volume of any primary THA
org/10.1136/bmjopen-2019-033045 recorded in the NJR in the preceding 365 days prior
to the index procedure in consenting patients i.e. time
Reproduced in summary form under CC BY varying volume.
4.0 licence.
Models were incrementally adjusted for patient
Background factors (age, gender, ASA grade, funder), operation
factors (fixation, approach, position, anaesthetic type,
Centralisation and specialisation in medical care are
thrombo-prophylaxis, bearing, year of surgery), centre
advocated to optimise a theorised volume-outcome
factors (private or public, centre surgical volume),
relationship i.e. higher volume surgeons and units are
consultant factors (training status, proportion of NHS
associated with better outcomes. Despite the prevailing
THA conducted in previous year, proportion of THAs
wisdom of such an association, evidence to support the
performed compared to all other joints), and deprivation
volume-outcome relationship in total hip arthroplasty
(English and Welsh IMD).
(THA) is sparse. In addition, investigating the volume-
outcome relationship is technically difficult due to the Statistical analysis was performed using a multi-level
computationally intensive methods required to calculate parametric (Weibull) survival model. Volume effect was
a time varying volume exposure. parametrised using restricted cubic splines. Analyses
were performed in Stata 15.1.
Differentiating between-consultant and within-
consultant effects is crucial to interpreting the data. Results
A between-consultant effect is essentially a cross-
sectional analysis that compares the performance of Of the 579,858 patients undergoing primary THA
one consultant against another and is highly likely to be (mean baseline age 69.8 years [SD 10.2]), 61.1% were
confounded by centre level effects. A within-consultant female. Figure 3.6 illustrates the distribution of within-
effect is based on individual time series data and consultant and between-consultant volume across the
compares changes of volume across time within the NJR. Figure 3.7 illustrates individual consultant and
same consultant. Correspondingly, within-consultant unit level variation in volume of procedures recorded
effects can be interpreted more strongly, as the effect by the NJR. Figures 3.8 and 3.9 show the results from
of changing a consultant’s personal volume, assuming multi-level survival models, they demonstrate differing
centre-level factors remain relatively constant over the results for between-consultant and within-consultant
short-term analysis period. The concept of between- effects. There was a strong volume-revision association
effect and within-effects is well known in epidemiology, between consultants (a cross-sectional association)
and often referred to as the ecological fallacy. with a near linear 43.3% (95% CI 29.1%-57.4%)
reduction of the risk of revision comparing consultants
The aim of this research is to investigate the between- with volumes between 1 and 200 procedures annually.
consultant and within-consultant (surgeon) effect of the Changes in individual surgeons (within-consultant) case
volume of primary THA for osteoarthritis (OA) and the volume showed no evidence of an association with
risk of subsequent revision. revision. Adjustment for confounding factors made little
difference to the reported associations.
Methods
Using data from the National Joint Registry of England,
Wales, Northern Ireland and the Isle of Man (NJR),
Figure 3.6 Empirical cumulative distribution and frequency distribution of (between) mean consultant
volume and (within) individual centred volume of hip arthroplasty in the previous 365 days. Grey
horizontal hashed lines indicate the 2.5th, 25th, 50th, 75th and 97.5th centiles of the distribution,
vertical hashed lines indicate mean and centred consultant volume at 2.5th, 25th, 50th, 75th and
97.5th centiles, respectively.
Between Consultant Volume

Mean Consultant Vol. of hip surgery in prev. 365 days
1
9 10
.1 .2 .3 .4 .5 .6 .7 .8 .9
Empirical Cumulative Distribution
8
7
Frequency (x100)
6
5
Frequency
4
3
2
1
0
0
0 100 200 300 400 500 600
Mean Consultant Volume of hip surgery in previous 365 days
Within Consultant Volume

Centred Consultant Vol. of hip surgery in prev. 365 days
1
250
.1 .2 .3 .4 .5 .6 .7 .8 .9
200
Frequency (x1000)
150

Frequency
100
50
0
-450 -350 -250 -150 -50 50 150

Centred Consultant Volume of hip surgery in previous 365 days
Figure 3.7 Mean, IQR and 95th centile range of consultant and centre volume of hip arthroplasty in
the previous 365 days recorded in the NJR by individual consultant and individual unit, respectively.
500
95th Centile range of volume recorded in NJR
IQR of volume recorded in NJR
Mean surgeon volume recorded in NJR
400
Consultant Volume
300
200
100
0
Consultant
95th Centile range of volume recorded in NJR
1500 IQR of volume recorded in NJR

Mean volume recorded in NJR
1200
Centre Volume
900
600
300
0
Centre
Figure 3.8 Between-consultant marginal association of hip surgical volume in the preceding 365 days
and hazard of revision arthroplasty unadjusted (M1) and adjusted (M5) for confounding factors in a
multilevel model (MLM). Patient factors include sex, American Society of Anesthesiologists grade and
funder. Operation confounding factors include fixation, approach, position, anaesthetic, mechanical
and chemical thromboprophylaxis, bearing and year of operation. Centre confounding factors include
hospital location and centre volume in the preceding 365 days. Surgeon confounding factors include
lead operating surgeon, listing of a surgeon within National Joint Registry prior to 2008, the proportion
of National Health Service cases in the preceding year and proportion hip arthroplasty procedures
undertaken in the previous year.
MLM Between M0:

Crude
1.2
1
Hazard Ratio
.8
.6
0 100 200 300 400 500 600
MLM Between M5:

MLM M4 + Adj Deprivation
1.2
1
Hazard Ratio
.8
.6
0 100 200 300 400 500 600
Consultant volume of hip surgery in previous 365 days
Figure 3.9 Within-consultant marginal association of hip surgical volume in the preceding 365 days
and hazard of revision arthroplasty unadjusted (M1) and adjusted (M5) for confounding factors in a
multilevel model (MLM). Patient factors include sex, American Society of Anesthesiologists grade and
funder. Operation confounding factors include fixation, approach, position, anaesthetic, mechanical
and chemical thromboprophylaxis, bearing and year of operation. Centre confounding factors include
hospital location and centre volume in the preceding 365 days. Surgeon confounding factors include
lead operating surgeon, listing of a surgeon within National Joint Registry prior to 2008, the proportion
of National Health Service cases in the preceding year and proportion hip arthroplasty procedures
undertaken in the previous year.
MLM Within M0:

1.3
Crude
1.2
Hazard Ratio
1.1
1
.9
-400 -300 -200 -100 0 100 200
MLM Within M5:

1.3
MLM M4 + Adj Deprivation

1.2
Hazard Ratio
1.1
1
.9
-400 -300 -200 -100 0 100 200
Consultant volume of hip surgery in previous 365 days
Discussion se. We suggest our analyses illustrate “state vs. trait”

behaviour. Where between-consultant association
We provide novel insights into the volume-outcome
illustrates the “traits” of surgeons, and within-consultant
relationship of 579,858 elective THA patients using a
associations illustrate their “state”. This is to say,
between-decomposition and within-decomposition
traits of experienced high-volume surgeons with
to analyse the association of consultant volumes on
good outcomes are unaffected by changes to their
revision. Uniquely, we use a time-varying volume
personal volume. Conversely, low volume arthroplasty
specification that facilitates the decomposition of
consultants who transiently increase their personal
between-consultant and within-consultant effects.
volume do not improve their outcomes.
We suggest the within-consultant effect is much
closer to the causal interpretation desired by many Conclusion
policymakers, and failure of research to recognise
the difference amongst between-effects and within- In summary, using data from the largest arthroplasty
effects may lead to erroneous policy decisions and register in the world, we have demonstrated that
unintended consequences. there is no within-consultant association between
surgical volume in the previous year and the risk of
We demonstrate that optimal between-consultant revision in patients undergoing primary THA for OA.
results are reached when the consultant volumes in Whereas there is strong evidence to suggest higher
the previous year are approximately 200 procedures. volume consultants tend to have better outcomes for
We suggest these factors are not causally related reasons that are unlikely to be due to the volume of
to volume, but rather due to unmeasured surgeon, arthroplasty performed by the individual consultant in
patient and/or centre factors. There is no evidence to the previous year per se.
suggest that consultants should change their personal
volume in the hope of improving their outcomes or that The results from this study have profound implications
there is an arbitrary threshold where the outcome of for quality improvement within healthcare. Encouraging
results become good. consultants to undertake a minimum number of
procedures under the guise of raising standards
We suggest the within-consultant effect from the could be counterproductive and may only serve to
multi-level regression is much closer to the causal expose patients to increased risk of revision by low or
interpretation required by consultants, patients, and previously low volume consultants. Centralisation and
policymakers i.e. what is the effect of changes in specialisation of THA in consultants who, for reasons
personal volume on the hazard of revision THA? This is not including volume, can undertake a greater number
not to say the between-effect is not of interest to policy of procedures is likely to benefit patients and reduce
makers, but to say that the between-effect suggests the revision burden overall. Encouraging or training low
that there are intrinsic differences between high and volume consultants to use prosthesis combinations
low volume consultants i.e. expertise, where higher with better outcomes may be a more effective method
volume consultants tend to have better outcomes, but of improving outcomes for patients.
these differences cannot be attributed to volume per
Published papers 2020-2021
Details of published analyses that have been sanctioned by the NJR Research Committee between April 2020
and March 2021. NJR data is available for research purposes following approval by the NJR Research Committee.
For further details please visit the NJR website at www.njrcentre.org.uk.
Patients Receiving a Primary Unicompartmental Knee Replacement Have a Higher Risk of Revision but
a Lower Risk of Mortality Than Predicted Had They Received a Total Knee Replacement: Data From the
National Joint Registry for England, Wales, Northern Ireland, and the Isle of Man.
Hunt LP, Blom AW, Matharu GS, Kunutsor SK, Beswick AD, Wilkinson JM, Whitehouse MR. J Arthroplasty. 2021
Feb;36(2):471-477 doi: 10.1016/j.arth.2020.08.063
Revision and 90-day mortality following hip arthroplasty in patients with inflammatory arthritis and
ankylosing spondylitis enrolled in the National Joint Registry for England and Wales.
Miller LL, Prieto-Alhambra D, Trela-Larsen L, Wilkinson JM, Clark EM, Blom AW, MacGregor AJ. Hip Int. 2021 Feb.
doi: 10.1177/1120700021990592
The Mid- to Long-Term Outcomes of the Lateral Domed Oxford Unicompartmental Knee Replacement:
An Analysis From the National Joint Registry for England, Wales, Northern Ireland, and the Isle of Man.
Mohammad HR, Matharu GS, Judge A, Murray DW. The J Arthroplasty. 2021 Jan;36(1):107-111. doi: 10.1016/j.
arth.2020.07.031
Choice between implants in knee replacement: protocol for a Bayesian network meta-analysis, analysis
of joint registries and economic decision model to determine the effectiveness and cost-effectiveness
of knee implants for NHS patients-The Knee Implant Prostheses Study (KNIPS).
Marques EMR, Dennis J, Beswick AD, Higgins J, Thom H, Welton N, Burston A, Hunt L, Whitehouse MR, Blom
AW. BMJ Open. 2021 Jan 6;11(1):e040205. doi: 10.1136/bmjopen-2020-040205
Similar risk of complete revision for infection with single-dose versus multiple-dose antibiotic
prophylaxis in primary arthroplasty of the hip and knee: results of an observational cohort study in the
Dutch Arthroplasty Register in 242,179 patients.
Veltman ES, Lenguerrand E, Moojen DJF, Whitehouse MR, Nelissen RGHH, Blom AW, Poolman RW. Acta Orthop.
2020 Dec;91(6):794-800. doi: 10.1080/17453674.2020.1794096
High mortality following revision hip arthroplasty for periprosthetic femoral fracture.
Khan T, Middleton R, Alvand A, Manktelow ARJ, Scammell BE, Ollivere BJ. Bone Joint J. 2020
Dec;102-B(12):1670-1674. doi: 10.1302/0301-620X.102B12.BJJ-2020-0367.R1
Personalized estimation of one-year mortality risk after elective hip or knee arthroplasty for
osteoarthritis.
Trela-Larsen L, Kroken G, Bartz-Johannessen C, Sayers A, Aram P, McCloskey E, Kadirkamanathan V, Blom AW,
Lie SA, Furnes ON, Wilkinson JM. Bone Joint Res. 2020 Nov;9(11):808-820. doi: 10.1302/2046-3758.911.BJR-
2020-0343.R1
What are the inpatient and day case costs following primary total hip replacement of patients treated
for prosthetic joint infection: a matched cohort study using linked data from the National Joint Registry
and Hospital Episode Statistics.
Garfield K, Noble S, Lenguerrand E, Whitehouse MR, Sayers A, Reed MR, Blom AW. BMC Med. 2020 Nov
18;18(1):335 doi: 10.1186/s12916-020-01803-7
Costs of joint replacement in osteoarthritis: a study using the National Joint Registry and Clinical
Practice Research Datalink datasets.
Leal J, Murphy J, Garriga C, Delmestri A, Rangan A, Price A, Carr A, Prieto-Alhambra D, Judge A. Arthritis Care
Res (Hoboken). 2020 Oct 1. doi: 10.1002/acr.24470
Provision of revision knee surgery and calculation of the effect of a network service reconfiguration: An
analysis from the National Joint Registry for England, Wales, Northern Ireland and the Isle of Man.
Kalson NS, Mathews JA, Miles J, Bloch BV, Price AJ, Phillips JRA, Toms AD, Baker PN; British Knee. 2020
Oct;27(5):1593-1600. doi: 10.1016/j.knee.2020.07.094
JointCalc: A web-based personalised patient decision support tool for joint replacement.
Zotov E, Hills AF, de Mello FL, Aram P, Sayers A, Blom AW, McCloskey EV, Wilkinson JM, Kadirkamanathan V. Int
J Med Inform. 2020 Oct;142:104217 doi: 10.1016/j.ijmedinf.2020.104217
A Comparison of the Surgical Practice of Potential Revision Outlier Joint Replacement Surgeons With
Non-outliers: A Case Control Study From the National Joint Registry for England, Wales, Northern
Ireland and the Isle of Man.
Penfold CM, Whitehouse MR, Sayers A, Wilkinson JM, Hunt L, Ben-Shlomo Y, Judge A, Blom AW. J Arthroplasty.
April 2021 [Epub ahead of print Oct 2020] doi: 10.1016/j.arth.2020.10.026
Association between surgical volume and failure of primary total hip replacement in England and
Wales: findings from a prospective national joint replacement register.
Sayers A, Steele F, Whitehouse MR, Price A, Ben-Shlomo Y, Blom AW. BMJ Open. 2020 Sep 14;10(9):e033045.
doi: 10.1136/bmjopen-2019-033045
A higher degree of polyethylene irradiation is associated with a reduced risk of revision for aseptic
loosening in total hip arthroplasties using cemented acetabular components: an analysis of 290,770
cases from the National Joint Registry of England, Wales, Northern Island and the Isle of Man.
Davis ET, Pagkalos J, Kopjar B. Bone Joint Res. 2020 Sep 20;9(9):563-571. doi: 10.1302/2046-3758.99.BJR-
2020-0135.R1
Risk Factors for Revision of Polished Taper-Slip Cemented Stems for Periprosthetic Femoral Fracture
After Primary Total Hip Replacement.
Lamb JN, Jain S, King SW, West RM, Pandit HG. J Bone Joint Surg Am. 2020 Sep 16;102(18):1600-1608. doi:
10.2106/JBJS.19.01242
Analysis of change in patient-reported outcome measures with floor and ceiling effects using the
multilevel Tobit model: a simulation study and an example from a National Joint Register using body
mass index and the Oxford Hip Score.
Sayers A, Whitehouse MR, Judge A, MacGregor AJ, Blom AW, Ben-Shlomo Y.BMJ Open. 2020 Aug
27;10(8):e033646. doi: 10.1136/bmjopen-2019-033646
Factors associated with implant survival following total hip replacement surgery: A registry study of
data from the National Joint Registry of England, Wales, Northern Ireland and the Isle of Man.
Evans JT, Blom AW, Timperley AJ, Dieppe P, Wilson MJ, Sayers A, Whitehouse MR PLoS Med. 2020 Aug
31;17(8):e1003291. doi: 10.1371/journal.pmed.1003291
The effect of surgical approach in total hip replacement on outcomes: an analysis of 723,904 elective
operations from the National Joint Registry for England, Wales, Northern Ireland and the Isle of Man.
Blom AW, Hunt LP, Matharu GS, Reed MR, Whitehouse MR. BMC Med. 2020 Aug 6;18(1):242. doi: 10.1186/
s12916-020-01672-0
A matched comparison of revision rates of cemented Oxford Unicompartmental Knee Replacements
with Single and Twin Peg femoral components, based on data from the National Joint Registry for
England, Wales, Northern Ireland and the Isle of Man.
Mohammad HR, Matharu GS, Judge A, Murray DW. Acta Orthop. 2020 Aug;91(4):420-425. doi:
10.1080/17453674.2020.1748288
Is aspirin as effective as the newer direct oral anticoagulants for venous thromboembolism prophylaxis
following total hip and knee replacement? An analysis from the National Joint Registry for England,
Wales, Northern Ireland and the Isle of Man.
Matharu G, Garriga C, Whitehouse N, Rangan A, Judge A. J Arthroplasty May 2020 Volume 35, Issue 9, 2631 -
2639.e6 doi: https://doi.org/10.1016/j.arth.2020.04.088
Effect of Bearing Surface on Survival of Cementless and Hybrid Total Hip Arthroplasty.
E, Pagkalos, J, Kopjar, B. JBJS Open Access: April-June 2020 - Volume 5 - Issue 2 - p e0075 doi: 10.2106/JBJS.
OA.19.00075
4. Implant and
unit-level activity
and outcomes
This section of the annual report gives performance
and data entry quality indicators for trusts and local 4.1 Implant performance
health boards (many of whom comprise more than
one hospital) and independent (private) providers in The NJR Implant Scrutiny Committee reports Level 1
England, Wales, Northern Ireland and the Isle of Man outlier implants to the MHRA. There are currently 11
for the 2020 calendar year. Outcomes analysis after hip stems, nine hip acetabular (cup) components and
hip and knee replacement surgery is also provided for 29 hip stem / cup combinations reported. A total of
the period 2011 to 2021. 14 knee brands are currently reported. Knee implants
with and without patella resurfacing are now included
This section also provides data for implant outliers in implant outlier analysis.
since 2003 and further information on notification and
last usage date. An implant is considered to be a Level 1 outlier when
its Prosthesis Time Incident Rate (PTIR) is more than
The full analysis for units can be found in the twice the PTIR of the group, allowing for confidence
document available in the downloads section at intervals. These are shown as the number of revisions
reports.njrcentre.org.uk per 100 prosthesis-years. As of March 2015, we have
started to identify the best performing implants, these
would have a PTIR less than half that of their group,
allowing for confidence intervals. To date no implants
have reached that level.
Components and constructs previously reported to

MHRA, but no longer at Level 1, are not listed.
Hip implant performance

Table 4.1 Level 1 outlier stems reported to MHRA.
Stem name Number implanted Latest PTIR Notified as outlier Last implanted
ASR 2,924 2.71 2010 2010
Corin Proxima* 105 2.11 2011 2009
S-ROM Cementless stem* 3,647 1.21 2013 Still in use

Adept Cementless stem* 228 1.89 2017 2010
Freeman Cementless 330 1.24 2019 2010
DePuy Proxima 341 1.31 2019 2014
Twinsys Cementless Stem 1,065 1.11 2019 2018
Alloclassic Cementless Stem 265 1.17 June 2020 Still in use
ESOP Stem 100 1.39 June 2020 2017
Bimetric Cementless Stem 4,947 0.87 February 2021 2019
SP II Cemented Revision 116 1.53 February 2021 Still in use
*Inclusion here is mainly due to metal-on-metal combinations.
Table 4.2 Level 1 outlier acetabular components reported to MHRA.
Cup name Number implanted Latest PTIR Notified as outlier Last implanted

ASR* 6,279 3.68 2010 2010
Ultima MoM cup* 194 1.68 2010 2006
R3 with metal liner** 151 2.88 2011 2011
M2A38* 1,489 1.67 2014 2011
Delta One TT 519 1.40 2015 Still in use
Trabecular Metal Revision Shell 419 1.43 2017 Still in use
seleXys TH+ 184 1.70 2018 2011
Pinnacle with metal liner** 15,601 1.32 2018 2013
MIHR cup* 257 1.82 2019 2011
*Inclusion here is mainly due to metal-on-metal combinations.

**Metal-on-metal.
Table 4.3 Level 1 outlier stem / cup combinations reported to MHRA.

Number Notified as Last
Combination implanted Latest PTIR outlier implanted
ASR Resurfacing Head / ASR Resurfacing Cup* 2,919 2.58 2010 2010
Metafix Stem / Cormet 2000 Resurfacing Cup* 173 2.50 2010 2011
CPT CoCr Stem / Adept Resurfacing Cup* 268 3.01 2011 2010
Corail / ASR Resurfacing Cup* 2,745 5.06 2011 2010
CPT CoCr Stem / BHR Resurfacing Cup* 117 2.42 2011 2010
Accolade / Mitch TRH Cup* 274 2.56 2011 2011
Summit Cementless Stem / ASR Resurfacing Cup* 128 4.42 2012 2009
CPT CoCr Stem / Durom Resurfacing Cup* 185 2.22 2012 2009
S-Rom Cementless Stem / ASR Resurfacing Cup* 148 3.66 2012 2010
CPCS / BHR Resurfacing Cup* 256 1.43 2012 2010
Anthology / BHR Resurfacing Cup* 513 2.71 2012 2011

SL-Plus Cementless Stem / Cormet 2000 Resurfacing Cup* 628 2.18 2013 2010
Profemur L Modular / Conserve Plus Resurfacing Cup* 164 2.40 2013 2010
Bimetric Cementless Stem / M2A 38* 1,303 1.70 2014 2011
Corin Proxima / Cormet 2000 Resurfacing Cup* 102 2.19 2015 2009
Synergy Cementless Stem / BHR Resurfacing Cup* 1,590 2.02 2016 2011
Adept Cementless Stem / Adept Resurfacing Cup* 201 2.01 2017 2010
Exeter V40 / Trabecular Metal Revision Shell 211 1.26 2017 2017
CLS Spotorno Cementless Stem / Adept Resurfacing Cup* 218 2.54 2017 2011
Spectron / Opera 216 1.06 2018 2014
Exeter V40 / Mitch TRH Cup* 121 1.54 2018 2010
Twinsys Cementless Stem / Adept Resurfacing Cup* 130 2.04 2018 2010
CLS Spotorno Cementless Stem / Durom Resurfacing Cup* 929 1.59 2018 2018
S-Rom Cementless Stem / Pinnacle* 2,085 1.24 2018 Still in use
S-Rom Cementless Stem / Ultima Mom Cup 105 1.46 2019 2005
Taperloc Cementless Stem / M2A 38* 138 1.47 2019 2010
Versys FMT Cementless Stem / Durom Resurfacing Cup* 182 1.43 2019 2010
Restoration Cementless Stem / Tritanium 119 2.87 June 2020 Still in use
Furlong HAC Stem / MIHR Cup 134 1.40 June 2020 2010
Best performing hip implants
There are no hip implants or combinations performing statistically less than half their expected PTIR.
Knee implant performance
Table 4.4 Level 1 outlier implants reported to MHRA.

Number
Knee brand implanted Latest PTIR Notified as outlier Last implanted
JRI Bicondylar Knee 248 1.68 2009 2008
Tack 231 1.57 2009 2008
St Leger 104 1.64 2011 2005
Journey Deuce 151 2.44 2014 2013

SLK Evo 103 1.72 2016 2013
ACS 203 1.57 2017 Still in use
Journey Oxinium 832 0.99 2017 2014
Smiles (METS hinged/linked knee)* 842 1.45 2018 Still in use
Endo-Model Modular Rotating Hinge* 245 2.04 2019 Still in use
Journey II BCS Oxinium without primary patella 713 1.50 June 2020 Still in use
E-Motion Bicondylar Knee with primary patella 329 1.24 June 2020 Still in use
Genesis II Oxinium without primary patella 5,346 0.83 February 2021 Still in use
LCS PFJ 223 4.81 February 2021 2010
RHK without primary patella 185 1.30 February 2021 2018
Genesis II Oxinium posterior stabilised 3,544 0.89 July 2021 Still in use
*Hinged knee prostheses are more often used in complex primaries, when compared to all total knee replacements.
Note: Analysis of knee replacements with and without patella resurfacing commenced in March 2020. Analysis by constraint (CR/PS/Constrained) commenced in
March 2021.
Best performing knee implants

There are no knee implants performing statistically less than half their expected PTIR.
4.2 Clinical activity resumed until July that year. It is expected that these
procedures will be identified by units and entered
retrospectively or submitted as part of our ongoing
Overall in 2020, 141 NHS trusts and local health audit programme.
boards (comprising 256 separate hospitals) and 177
independent hospitals were open and eligible to report Compliance is measured by comparing the proportion
patient procedures to the registry. Data were not of all joint replacements entered into the registry, with
submitted in 2020 by five NHS hospitals (including those submitted to the Hospital Episodes Statistics
two trauma units) and two independent hospitals. Due (HES) and Patient Episode Database Wales (PEDW)
to the COVID-19 pandemic, hospitals were directed services. These data rely on submissions by hospitals
at the end of March 2020 to cease inputting records and are only available by NHS trust. No data are
to audits and to divert staff to higher priority tasks, currently available from private providers and figures
therefore a number of elective procedures were not also exclude units in Northern Ireland as compliance
initially reported as data entry activity wasn’t then data are not available.
• 37% of NHS hospitals in England and Wales Independent hospitals

reported 95% or more of the joint replacements they
• 68% achieved a proportion of patients with a linkable
undertook
NHS number greater than 95%
• 37% reported between 80% to 95%
• 25% achieved a proportion of 80% to 95%
• 26% reported less than 80%
• 7% recorded a proportion of less than 80%
Of those hospitals submitting data, the proportion
There has been a drop in linkability from 2019, with the
of patients who gave permission (consent) for their
percentage of submitting units achieving over 95% in
details to be entered into the registry were:
2020 falling from 80% to 76%. The proportion achieving
NHS hospitals a greater than 80% linkability rate has increased slightly
to 20% compared with 17% in 2019. The drop in
• 33% of NHS hospitals achieved a consent rate of linkability is related to the fall in consent rate.
greater than 95%
• 36% achieved a consent rate of 80% to 95% Note: Independent hospitals might be expected to
have lower linkability rates than NHS hospitals, as a
• 31% recorded a consent rate of less than 80%
proportion of their patients may come from overseas
Independent hospitals and do not have an NHS number.
• 57% of independent hospitals achieved a consent

rate greater than 95% 4.3 Outlier units for
• 30% achieved a consent rate of 80% to 95%
• 13% recorded a consent rate of less than 80%
90-day mortality and
There has been a decrease in recorded consent for revision rates for the
all submitting units when compared to the previous
year, with those achieving a higher than 95% rate period 2011 to 2021
falling to 43% from 56% in 2019. The proportion of
all units achieving a higher than 80% consent rate, The observed numbers of revisions of hip and knee
has increased slightly. This reduction in consent rate replacements for each hospital were compared to
can be related to the ratio of elective to trauma cases, the numbers expected, given the unit’s case-mix
which changed significantly during 2020, having in respect of age, gender and reason for primary
a higher proportion of trauma cases compared to surgery. Hospitals with a much higher than expected
previous years. There was a significant reduction in revision rate for hip and knee replacement have been
elective cases due to COVID-19 and trauma cases identified. These hospitals had a revision rate that was
have a higher rate where NJR consent is not obtained. above the upper of the 99.8% control limits (these
limits approximate to +/-3 standard deviations). We
Similarly, the proportion of entries in which there is would expect 0.2% (i.e. one in 500) to lie outside the
significant data to enable the patient to be linked to an control limits by chance, with approximately half of
NHS number (linkability) is listed. these (one in 1,000) to be above the upper limit.
NHS hospitals When examined over the past ten years of the
registry, a total of 34 hospitals reported higher than
• 81% achieved a proportion of patients with a linkable
expected rates of revision for knee replacement, and
NHS number greater than 95%
23 hospitals had higher than expected rates of revision
• 17% achieved a proportion of 80% to 95% for hip surgery. However, revisions taken only from the
• 2% recorded a proportion of less than 80% last five years of the registry showed only 13 hospitals
reporting higher than expected rates for knees, and 11
for hips.
The 90-day mortality rate for primary hip and knee Table 4.7 Outliers for hip revision rates, all linked
replacement was calculated using the last five years of primaries from 20111.
data for all hospitals by plotting standardised mortality
ratios for each hospital against the expected number Hospital name
of deaths. No hospitals had higher than expected BMI Clementine Churchill Hospital (Middlesex)
mortality rates for either hip or knee replacement. BMI The Meriden Hospital (West Midlands)
Bradford Royal Infirmary
Note: The case mix for mortality includes age, gender
Broadgreen Hospital
and ASA grade. Trauma cases have been excluded
Chorley and South Ribble Hospital
from both the hip and knee mortality analyses together
Colchester General Hospital
with hips implanted for failed hemi-arthroplasty or for
metastatic cancer (the latter only from November 2014 Fitzwilliam Hospital (Cambridgeshire)
when recording of this reason began). Also, where Homerton University Hospital
both left and right side joints were implanted on the Hospital of St Cross
same day, only one side was included in the analysis. Milton Keynes Hospital
North Downs Hospital (Surrey)
Note: Any units identified as potential outliers here Nuffield Orthopaedic Centre
have been notified. All units are provided with an NJR Orthopaedics and Spine Specialist Hospital
Annual Clinical Report and additionally have access to (Cambridgeshire)
the online NJR Management Feedback system. Salisbury District Hospital
South Tyneside District Hospital
Important note about the outlier hospitals listed
Southampton General Hospital
In earlier annual reports, we reported outlying hospitals Spire Hartswood Hospital (Essex)
based on all cases submitted to the registry since 1 St Richard's Hospital
April 2003. To reflect changes in hospital practices Sussex Orthopaedic NHS Treatment Centre
and component use, we now report outlying hospitals The Tunbridge Wells Hospital
based on the last ten years (13 February 2011 to 13
Wansbeck Hospital
February 2021) and five years of data (13 February
Watford General Hospital
2016 to 13 February 2021 inclusive, the latter date
Weston General Hospital
being when the dataset was cut). These cuts of data
exclude the majority of withdrawn outlier implants and
metal-on-metal total hip replacements from analysis,
and thus better represent contemporary practice. Table 4.8 Outliers for hip revision rates, all linked
primaries from 20162.
Table 4.5 Outliers for hip mortality rates since 20162. Hospital name
Castle Hill Hospital
Hospital name
Clifton Park Hospital (North Yorkshire)
None identified
Darent Valley
Fulwood Hall Hospital (Lancashire)
Hexham General Hospital
Table 4.6 Outliers for knee mortality rates since 20162.
Milton Keynes Hospital
North Tyneside General Hospital
Hospital name
Nuffield Orthopaedic Centre
None identified
Wansbeck Hospital
Weston General Hospital
Table 4.9 Outliers for knee revision rates, all linked Table 4.10 Outliers for knee revision rates, all linked
primaries from 20111. primaries from 20162.
Hospital name Hospital name

Abergele Hospital BMI Bath Clinic (Avon)
Ashford Hospital BMI The South Cheshire Private Hospital (Cheshire)
BMI Bishops Wood Hospital (Middlesex) Guy's Hospital
BMI Goring Hall Hospital (West Sussex) Hillingdon Hospital
BMI The London Independent Hospital (Greater London) King Edward VII’s Hospital Sister Agnes (Greater London)
BMI The Meriden Hospital (West Midlands) Nuffield Orthopaedic Centre
Broadgreen Hospital Practice Plus Group Hospital - Barlborough (Derbyshire)
County Hospital Louth Queen Elizabeth The Queen Mother Hospital
Ealing Hospital Southmead Hospital
Guy's Hospital Spire Bushey Hospital (Hertfordshire)
Heatherwood Hospital Springfield Hospital (Essex)
Hillingdon Hospital The Royal National Orthopaedic Hospital (Stanmore)
Hinchingbrooke Hospital Yeovil District Hospital
King Edward VII’s Hospital Sister Agnes (Greater London)
Mount Vernon Treatment Centre Note: 1 Date range 13 February 2011 to 13 February 2021 inclusive. 2 Date
range 13 February 2016 to 13 February 2021 inclusive.
Nevill Hall Hospital
Nuffield Health Chichester Hospital (West Sussex)
Nuffield Orthopaedic Centre
Orthopaedics and Spine Specialist Hospital
(Cambridgeshire)
Peterborough City Hospital
Queen Elizabeth The Queen Mother Hospital
Southmead Hospital
Spire Hull and East Riding Hospital (East Yorkshire)
Spire Southampton Hospital (Hampshire)
Springfield Hospital (Essex)
St Mary's Hospital (Isle of Wight)
St Richard's Hospital
Sussex Orthopaedic NHS Treatment Centre
The Royal National Orthopaedic Hospital (Stanmore)
Torbay Hospital
University College Hospital
University Hospital Llandough
York Hospital
4.4 Better than expected Table 4.13 Better than expected knee revision rates,
all linked primaries from 20111.
performance Hospital name

Bishop Auckland Hospital
This year we have again listed hospitals where revision Burnley General Hospital
rates are statistically better than expected. The lists
Claremont Hospital (South Yorkshire)
here show units that lie below the 99.8% control limit
Craigavon Area Hospital
which also achieved greater than 90% compliance
Hexham General Hospital
across all of the NJR data quality audits. Units with
lower data quality compliance are automatically Ipswich Hospital
excluded from these lists. Musgrave Park Hospital
Norfolk and Norwich Hospital
North Tyneside General Hospital
Table 4.11 Better than expected hip revision rates, Nottingham Woodthorpe Hospital (Nottinghamshire)
all linked primaries from 20111. Nuffield Health Cambridge Hospital (Cambridgeshire)
Nuffield Health Derby Hospital (Derbyshire)
Hospital name Nuffield Health Ipswich Hospital (Suffolk)
Calderdale Royal Hospital Princess Alexandra Hospital
Ipswich Hospital Spire Norwich Hospital (Norfolk)
Musgrave Park Hospital Stepping Hill Hospital
Royal Derby Hospital The Elective Orthopaedic Centre
Royal Devon and Exeter Hospital (Wonford)
Royal Surrey County Hospital
Sunderland Royal Hospital Table 4.14 Better than expected knee revision rates,
all linked primaries from 20162.
Hospital name
Table 4.12 Better than expected hip revision rates, all
Ipswich Hospital
linked primaries from 20162.
Musgrave Park Hospital
Nottingham Woodthorpe Hospital (Nottinghamshire)
Hospital name
Calderdale Royal Hospital
Note: 1 Date range 13 February 2011 to 13 February 2021 inclusive. 2 Date
Musgrave Park Hospital
range 13 February 2016 to 13 February 2021 inclusive.
Ulster Independent Clinic (Belfast)
The effects of
the COVID-19
pandemic on
joint replacement
surgery volumes
and waiting lists
The COVID-19 induced
joint replacement deficit
in England, Wales and
Northern Ireland
Adrian Sayers PhD,1 Kevin Deere MSc,1 Erik Lenguerrand PhD,1 Setor K Kunutsor PhD,1+2 Jonathan L Rees MD,3+9
Andy Judge PhD,1 Yoav Ben-Shlomo PhD,1+4 Celia L Gregson PhD,1 Emma M Clark MD, PhD,1 Mike Reed MD,5
Timothy Wilton MA,6 Derek J Pegg FRCS Ed(Tr&Orth),7 Jeremy M Wilkinson PhD,8 Andrew Price DPhil,3+9
Michael R Whitehouse PhD,1+2 Ashley W. Blom PhD.1+2
1 Musculoskeletal Research Unit, Translational Health Sciences, Bristol Medical School, 1st Floor Learning &
Research Building, Southmead Hospital, Bristol, BS10 5NB, UK
2 National Institute for Health Research Bristol Biomedical Research Centre, University Hospitals Bristol and
Weston NHS Foundation Trust and University of Bristol
3 Nuffield Department of Orthopaedics, Rheumatology and Musculoskeletal Sciences, Botnar Research Centre,
University of Oxford, Oxford, OX3 7LD, UK
4 Population Health Sciences, University of Bristol, Bristol BS8 2PS
5 Northumbria Healthcare NHS Foundation Trust, Department of Trauma and Orthopaedics, Wansbeck General
Hospital, Woodhorn Lane, NE63 9JJ, UK
6 Department of Orthopaedics, Royal Derby Hospital, Uttoxeter Rd, Derby, DE22 3NE, UK
7 Mid Cheshire Hospitals NHS Foundation Trust, Leighton Hospital, Middlewich Road, Crewe, Cheshire, CW1 4QJ
8 Department of Oncology and Metabolism, The University of Sheffield, Sorby Wing, Northern General Hospital,
Sheffield, S5 7AU, UK
9 National Institute for Health Research Oxford Biomedical Research Centre, Oxford University Hospitals
Foundation Trust
This work is licensed under a

Creative Commons Attribution 4.0 International License
https://creativecommons.org/licenses/by-nc/4.0
National Joint Registry | The effects of the COVID-19 pandemic on joint replacement surgery
Summary COVID-19. As the pandemic evolves, further waves

of infection are likely to restrict surgery and see the
Background deficit increase, therefore projections of time taken
to address the deficit must thus be regarded as the
The COVID-19 pandemic has impacted health, best-case scenario. A significant expansion of joint
economies and the functioning of societies globally. replacement services compared to 2019 is urgently
In addition to direct health effects, it has indirectly required to address this deficit.
impacted population health by limiting access to
non-COVID treatments, including joint replacements. Funding
The pandemic has necessitated re-organisation of
This study was supported by the NIHR Biomedical
healthcare with the private-sector providing support
Research Centre at University Hospitals Bristol and
to public hospitals in some areas. The full impact is
Weston NHS Foundation Trust, the University of
therefore difficult to ascertain from public data
Bristol, the NIHR Biomedical Research Centre at
sources alone.
Oxford University Hospitals NHS Foundation Trust
and the University of Oxford. The views expressed are
Methods
those of the author(s) and not necessarily those of the
We used a mandatory prospective national register NIHR or the Department of Health and Social Care.
of private and publicly funded hip, knee, shoulder, This study was also supported by funding from the
elbow and ankle replacements in England, Wales and National Joint Registry. Employment posts of authors
Northern Ireland. Descriptive analysis of the provision of this work are part-funded by a grant from the
of joint replacement comparing data from 2019 to National Joint Registry to conduct statistical analysis
2020 and predicted deficit recovery. for the National Joint Registry.
Findings Supplementary Material

There was a substantial deficit in the provision of joint Additional graphs and tables mentioned in this paper
replacement in 2020 compared to 2019 with 106,922 can be found online either by scanning the following
(48.8%) fewer procedures performed; resulting in QR Code or visiting the link below:
45,116 (44%) fewer hip replacements, 57,115 (52%)
fewer knee replacements, 3,878 (50%) fewer shoulder https://reports.njrcentre.org.uk/COVID19
replacements, 280 (33%) fewer elbow replacements
and 533 (53%) fewer ankle replacements performed.
Wales and Northern Ireland were disproportionately
affected with an overall reduction of 8,001 (67%) and
2,833 (64%) respectively compared to 96,088 (47%) in
England.
An immediate 5% expansion in provision from

the 2019 baseline will eliminate the deficit over
approximately 10 years (by 2031), whilst a 10%
expansion will address the deficit by 2026.
Interpretation
This large national analysis of both private and publicly
funded joint replacements illustrates a substantial
accumulated deficit of surgery, equivalent to six
months of normal activity across England, Wales
and Northern Ireland, due to the indirect effects of
Introduction private hospitals by NHS trusts. The analysis of single
centres, public sector or private sector databases may
Joint replacement is a common and important surgical be misleading as they are unable to consider the totality
procedure used to treat a variety of musculoskeletal of a healthcare system that has become increasingly
problems including osteoarthritis and acute trauma. integrated during this pandemic. A comprehensive
It is a highly successful procedure that reduces pain analysis of both private and public sector provision
and disability enabling participation in and contribution is required to understand the impact of COVID-19
to society. The Lancet described joint replacement and plan the recovery of joint replacement capacity.
as the operation of the 20th century.1 Over 200,000 Fortunately, England, Wales and Northern Ireland have
primary hip and knee replacements were performed an integrated mandatory register, the National Joint
in England, Wales and Northern Ireland in 2019.2 Joint Registry, for all hip, knee, shoulder, elbow and ankle
replacements are long-lasting, with over half of hip and joint replacements.
knee replacements lasting in excess of 25 years3,4 and
We aim to describe the impact of the COVID-19
90% of shoulder replacements lasting in excess of 10
pandemic on joint replacement services in England,
years.5 For very many people it is a curative procedure
Wales and Northern Ireland and quantify the
for the debilitating effects of end-stage arthritis.
expansion of services required in order to address the
The COVID-19 pandemic has had an unprecedented
accumulated deficit of joint replacement surgery and
impact on populations around the world. The first
return joint replacement service provision back to pre-
patient with COVID-19 in the UK was identified on the
pandemic levels.
23rd January 20206 and the first UK national lockdown
commenced on 23rd March 2020.7 The pandemic has
impacted our lives widely and has inevitably required
Methods
a massive and rapid re-organisation of healthcare Data Source
provision in order to provide care for patients with
severe acute respiratory distress due to SARS-Cov2 In this prospective observational registry-based study
infection. we analysed data from the National Joint Registry
(NJR).2 We collected data on hip, knee, shoulder, elbow
Less urgent medical procedures have been forgone and ankle primary joint replacement procedures entered
or deferred due to competing pandemic demands. into the registry from hospitals in England, Wales
We have seen a re-organisation of services from the and Northern Ireland since its inception in April 2003
public to the private sector and some hospitals have through to the end of December 2020. A data quality
specialised in “COVID care”, whereas others have audit in 2017/18 showed over a 95% and 96% capture
attempted to remain “COVID free” in an effort to provide of all primary hip and knee data respectively (though
more routine services. Early reports have suggested this has subsequently been significantly improved by
that mortality is persistently high (20.4%) following the national audits).2
acquisition of COVID-19 in the perioperative period
after elective surgery.8 Reports from around the world The NJR data was prepared for this analysis in the
have suggested a wide variety of consequences of same manner as described in the NJR 2020 17th
healthcare reorganisation including a reduction in the Annual Report.2 Data were cleaned by removing
volume of joint replacement,9-13 an increased number records with missing information, removing duplicate
of patients with symptoms “worse than death” whilst procedures, and removing records where we were
waiting for joint replacement,14 increased waiting lists15 unable to ascertain a logical sequence of revision
and economic hardship.16 However, the majority of procedures. The cleaning process resulted in 2,789,980
these reports have been based on single centres primary procedures for analysis (see Supplementary
with small sample sizes. Assessing the impact of Figures 1 to 5, see online for all supplementary figures).
COVID-19 on the provision of joint replacement is
difficult due to the shift in surgeries from the public to
Statistical Analysis
the private sector and the effective commandeering of We used descriptive statistics to illustrate the impact
of COVID-19 on the provision of joint replacement

since the start of data collection for each type of joint
Results
replacement, dividing procedures into acute (those Overall, 106,922 (48.8%) fewer joint (hip, knee,
performed for trauma) and elective indications, shoulder, elbow, ankle) replacements were performed
where possible. in 2020 compared to 2019. Knee replacements
showed the largest reduction in absolute numbers
We present weekly counts of procedures in 2019
followed by hip replacements, see Table 1. Wales and
compared to 2020 by each joint, dividing procedures
Northern Ireland have recorded 67% and 64% fewer
into acute and elective indications where possible, and
joint replacement procedures respectively compared
include a 21-day weekly rolling average.
to 2019, which is substantially greater than the deficit
The time-to-recovery and expansion in services of 47% experienced by England.
required compared to 2019 was also calculated. We
assume the years-to-recovery is estimated by deficit
in procedures expressed as a percentage expansion
of services compared to 2019 i.e. a 50,000 procedure
deficit will take 5 years to recover assuming a baseline
provision of 100,000 patients and a 10% expansion in
surgical provision. We have simplistically assumed a
static baseline (2019) though the secular patterns prior
to this suggest the need for increasing service provision
(with the possible exception of knee replacement) so
these estimates are likely to be conservative.
Time-to-recovery was calculated for England,

Wales and Northern Ireland overall and for each
nation separately.
Sensitive Analysis
Weekly frequencies were also calculated for all English
sub-regions for all joints in 2020, dividing procedures
into acute and elective indications where possible.
Time-to-recovery was also calculated for all English
sub-regions for all joints comparing provision to 2019.
All analyses were conducted in Stata 15.1 StataCorp.
College Station, TX.
Role of the funding source

The funder of the study had no role in study design,
data collection, data analysis, data interpretation, or
writing of the report. AS, KD, EL had full access to
all the data in the study and all authors had the final
responsibility for the decision to submit for publication.
Table 1: Descriptive statistics of provision and change of joint replacement by joint and nation.
Joint N(2019) N(2020) N(Change) %(Change)

England, Wales and Northern Ireland
Hip 100,940 55,824 -45,116 -44.7
Knee 108,607 51,492 -57,115 -52.6
Shoulder 7,737 3,859 -3,878 -50.1
Elbow 850 570 -280 -32.9
Ankle 1,009 476 -533 -52.8
Total 219,143 112,221 -106,922 -48.8
England
Hip 93,148 52,818 -40,330 -43.3
Knee 100,547 49,169 -51,378 -51.1
Shoulder 7,373 3,736 -3,637 -49.3
Elbow 791 552 -239 -30.2
Ankle 961 457 -504 -52.4
Total 202,820 106,732 -96,088 -47.4
Wales
Hip 5,501 2,039 -3,462 -62.9
Knee 6,025 1,734 -4,291 -71.2
Shoulder 287 97 -190 -66.2
Elbow 43 9 -34 -79.1
Ankle 32 8 -24 -75.0
Total 11,888 3,887 -8,001 -67.3
Northern Ireland
Hip 2,291 967 -1,324 -57.8
Knee 2,035 589 -1,446 -71.1
Shoulder 77 26 -51 -66.2
Elbow 16 9 -7 -43.8
Ankle 16 11 -5 -31.3
Total 4,435 1,602 -2,833 -63.9
Figure 1 illustrates the difference in accrual of primary

joint replacements since the start of data collection for
each joint. Data have illustrated that provision of joint
replacement has increased year on year since data
collection started.
Figure 1: Annual number of primary hip, knee, shoulder, elbow and ankle replacements performed in England,
Wales and Northern Ireland.
Hip Knee
100,000
100,000
80,000
80,000
Primary procedures
Primary procedures
60,000
60,000
40,000
40,000
20,000 20,000
0 0
2003 2005 2007 2009 2011 2013 2015 2017 2019 2003 2005 2007 2009 2011 2013 2015 2017 2019
2004 2006 2008 2010 2012 2014 2016 2018 2020 2004 2006 2008 2010 2012 2014 2016 2018 2020
Shoulder Elbow
8,000 1,000
900
7,000
800
6,000
700
Primary procedures
Primary procedures
5,000
600
4,000 500
400
3,000
300
2,000
200
1,000
100
000,000 000,000
2012 2013 2014 2015 2016 2017 2018 2019 2020 2012 2013 2014 2015 2016 2017 2018 2019 2020
Ankle
1,000
900
800
700
Primary procedures
600 Acute trauma

500
400 Elective
300
200
100
000,000
2010 2012 2014 2016 2018 2020
2011 2013 2015 2017 2019
Figures 2 and 3 illustrate weekly counts of primary first eight weeks following the first national lockdown.
hip, knee, shoulder, elbow and ankle replacements The volumes of acute procedures (those performed
in 2019 compared to 2020. These show a rapid for trauma) recorded in hip, shoulder and elbow
decline in the number of procedures prior to the start replacements were also reduced in 2020 compared
of the first national lockdown. A very small number to 2019.
of elective joint replacements were performed in the
Figure 2: Weekly number of primary hip and knee replacements performed in England, Wales, and Northern
Ireland in 2019 and 2020.
Hip
2019 2020
2500
2000
Primary procedures
1500
1000
500
0
Jan Dec Jan Dec
Knee
2019 2020
2500
2000
Primary procedures
1500
1000
500
0
Jan Dec Jan Dec
Acute trauma Elective Weekly rolling average
Graphs by year of primary operation

Red line indicates first national lockdown
Weekly (centred) rolling average based over 21 days
Figure 3: Weekly number of primary shoulder, elbow and ankle replacements performed in England, Wales, and
Northern Ireland in 2019 and 2020.
Shoulder
2019 2020
200
150
Primary procedures
100
50
0
Jan Dec Jan Dec
Elbow
2019 2020
30
Primary procedures
20
10
0
Jan Dec Jan Dec
Ankle
2019 2020
30
Primary procedures
20
10
0
Jan Dec Jan Dec
Graphs by year of primary operation

Red line indicates first national lockdown
Weekly (centred) rolling average based over 21 days
Figure 4 shows a breakdown of weekly counts in Wales and Northern Ireland in the second quarter
of primary hip, knee, shoulder, elbow and ankle of the year (2020) is negligible compared to those
replacements in 2020 stratified by each nation. This recorded in England. This pattern is even more
shows that the reduced volume of joint replacements pronounced for shoulder, elbow and ankle procedures
is not evenly distributed across England, Wales and in the 2nd, 3rd and 4th quarters of 2020.
Northern Ireland. The volume of procedures recorded
Figure 4: Weekly number of primary shoulder, elbow and ankle replacements performed in England,Wales, and
Northern Ireland in 2020 by nation.
2000 England 150 Wales 75 Northern Ireland
100 50
1000
Hip
50 25
0000
0 000
0 00
0
Jan Dec Jan Dec Jan Dec
England 150 Wales 50 Northern Ireland

2000
100
Knee
1000 25
50
0000
0 000
0 00
0
Primary procedures
150 15 15
Shoulder
100 10 10
50 5 5
0000
0 000
0 00
0
20
Elbow
10
0000
0 000
0 00
0
20
Ankle
10
0000
0 000
0 00
0
Supplementary Figures 6 through to 10 and Figure 5 illustrates the years-to-recovery following

Supplementary Table 1 illustrate regional breakdown expansion of provision compared to 2019 rates across
of weekly counts of primary hip, knee, shoulder, elbow England, Wales and Northern Ireland stratified by joint.
and ankle replacements in 2020. Supplementary This figure illustrates that an immediate 5% expansion
Figure 6 and 7 demonstrate heterogeneity in the in provision of hip, knee, shoulder, elbow and ankle
recovery of hip and knee replacements from the first replacement compared to 2019 may address the
wave of COVID-19 infections, with some regions deficit in procedures, within approximately 10 years.
beginning restoring provision more rapidly, and to a A 10% expansion in provision is projected to address
greater extent, than others. the current deficit in approximately five years.
Figure 5: Predicted years-to-recovery of the 2020 deficit of joint replacement procedures following expansion of
joint replacement provision compared to 2019 in England, Wales and Northern Ireland.
50
40
Years to recovery
30
20
10
0
0 5 10 15 20 25 30
% Expansion in service compared to 2019
Hips Knees Shoulders Elbows Ankles
Figure 6 illustrates the years-to-recovery following

expansion of provision compared to 2019 stratified by
England, Wales and Northern Ireland and joint type.
Figure 6 and data in Table 2 illustrate that the recovery
in Wales and Northern Ireland will take longer for an
equivalent expansion in services.
Figure 6: Predicted years-to-recovery of the 2020 deficit of joint replacement procedures following expansion of
joint replacement provision compared to 2019 stratified by nation.
80 England 80 Wales
70 70
60 60
50 50
40 40
30 30
20 20
10 10
Years to recovery
0 0
0 5 10 15 0 5 10 15
80 Northern Ireland
70
60
50
40
30
20
10
0
0 5 10 15
Hips Knees Shoulders Elbows Ankles
% Expansion in service compared to 2019
Supplementary Figure 11 and Supplementary Table

2 illustrates years-to-recovery following expansion
of provision compared to 2019 stratified by region
and joint type. These data illustrate heterogeneity
in provision of joint replacement during 2020 and
different recovery profiles in the 3rd and 4th quarters
of 2020.
Table 2: Predicted years-to-recovery of 2020 deficit following expansion of joint replacement provision compared
to 2019 by joint type and nation.
Expansion compared Years to recovery

to 2019 (%) Hip Knee Shoulder Elbow Ankle
England, Wales and Northern Ireland
5 8.9 10.5 10.0 6.6 10.6
10 4.5 5.3 5.0 3.3 5.3
15 3.0 3.5 3.3 2.2 3.5
20 2.2 2.6 2.5 1.6 2.6
25 1.8 2.1 2.0 1.3 2.1
30 1.5 1.8 1.7 1.1 1.8
England
5 8.7 10.2 9.9 6.0 10.5
10 4.3 5.1 4.9 3.0 5.2
15 2.9 3.4 3.3 2.0 3.5
20 2.2 2.6 2.5 1.5 2.6
25 1.7 2.0 2.0 1.2 2.1
30 1.4 1.7 1.6 1.0 1.7
Wales
5 12.6 14.2 13.2 15.8 15.0
10 6.3 7.1 6.6 7.9 7.5
15 4.2 4.7 4.4 5.3 5.0
20 3.1 3.6 3.3 4.0 3.8
25 2.5 2.8 2.6 3.2 3.0
30 2.1 2.4 2.2 2.6 2.5
Northern Ireland
5 11.6 14.2 13.2 8.8 6.3
10 5.8 7.1 6.6 4.4 3.1
15 3.9 4.7 4.4 2.9 2.1
20 2.9 3.6 3.3 2.2 1.6
25 2.3 2.8 2.6 1.8 1.3
30 1.9 2.4 2.2 1.5 1.0
Discussion volume treatment centres each providing 500 hip
and 500 knee joint replacements per year. Staffing
We present the first comprehensive assessment of such facilities and providing all the ancillary care
the provision of joint replacement across the entire would also be extremely challenging. Any additional
health service (private and public) in England, Wales theatre capacity developed will require consultant
and Northern Ireland. The COVID-19 pandemic has orthopaedic surgeons, anaesthetic staff, theatre staff,
had a profound impact on patients due to reduced nurses, physiotherapists and all the other ancillary
service delivery of joint replacement surgery. Provision services. Expanding staff capacity cannot take place
of joint replacement surgery in 2020 was reduced overnight and presents the most serious challenge.
by approximately 50% compared to 2019. Patients Utilisation and efficiency solutions are likely to offer
requiring elective joint replacement have been only a partial answer. Caution must also be exercised
impacted the most with acute trauma provision being when attempting to expand capacity within existing
largely preserved throughout 2020. The impact of staff, ensuring they are retained and supported in
COVID-19 has not been uniform across or within the order that work-related “burn-out” due to COVID-19
nations covered by the NJR. Wales and Northern is not exacerbated. Similarly, the increased volume
Ireland have seen the greatest reduction in capacity of post discharge care will have significant resource
with surgery for patients requiring elective shoulder, implications and impact on already stretched
elbow and ankle replacements effectively being halted. community-based services.
We illustrate that to recover the accumulated deficit The removal of barriers to increasing capacity, such
in joint replacement that has occurred in 2020 as annual17 and lifetime pension allowances,18 will be
a significant expansion in pre-pandemic service as important as incentivising 7-day a week operating,
provision is needed, even if it is assumed that demand asking senior orthopaedic surgeons and anaesthetists
remains static at 2019 levels, which is unlikely to be to delay their retirement or asking recently retired
true given the year-on-year secular increase in the surgeons and anaesthetists to return from retirement
provision of most procedures except possibly knee to assist in the provision of joint replacement are all
replacement. The deficit in 2020 is equivalent to six strategies to be considered. Expanding bed capacity
months of normal activity across England, Wales and will be particularly difficult during winter months when,
Northern Ireland. Without expansion in provision, even prior to the pandemic, elective surgery is already
waiting lists for joint replacement will be, at a minimum routinely curtailed. Minimising seasonal variation in
six months longer compared to pre-pandemic levels the volume of primary procedures performed will
based on the assumption that services have been be essential in maximising service delivery; the role
restored since January 2021. However, as provision of private sector service provision is likely to be
had not recovered to pre-pandemic levels by the end increasingly important in restoring provision of joint
of 2020, it is likely that the pandemic will continue replacement.
to impact patients due to reduced provision of joint
The selection of evidence based joint replacement19,20
replacement services for at least the first half of 2021.
and rehabilitation strategies which are cost-effective21
Waiting lists will therefore continue to lengthen as the
and minimise the national healthcare and revision
deficit increases.
burden will be essential in maximising capacity of
Expanding provision in the post-pandemic NHS primary procedures, with NHS initiatives such as
system will be challenging. Either greater productivity, “Getting It Right First Time”22 playing an important
equivalent to every hospital providing an additional 2.5 role. The rapid assessment of the clinical and cost-
or 5 weeks of joint replacement provision per year, effectiveness of new treatment modalities, such as
must be achieved which is unlikely to be feasible. An day case joint replacement,23 and enhanced recovery
alternative strategy would be a 5% or 10% expansion programmes24 are required if the capacity for primary
in services crudely represents 10 or 20 new high joint replacement is to be maximised.
Strengths and limitations Conclusion

This analysis has a number of strengths. Importantly
We have been able to reliably assess the impact
the data included in this analysis covers both private
on patients waiting for joint replacement, created
and publicly-funded joint replacement procedures.
by the effects of COVID-19 in 2020, using
Contribution to the registry is mandatory and primary
a nationally representative data source. The
case ascertainment is in excess of 95% for hip and
provision of primary joint replacement declined
knee replacements.2
by approximately 106,000 cases (50%) in 2020
We assume the latent demand for joint replacement in England, Wales and Northern Ireland. This
will be the same as 2019, we have not accounted will inevitably lead to a large number of patients
for the increased demand in joint replacement we enduring unnecessary pain, disability and secondary
have seen historically, which is approximately 5% decline in mental and overall physical well-being.
per year.2 We have not factored in the reduced
The impact on waiting times, in an already
demand for joint replacement due to the higher
overstretched healthcare system, is extremely
expected mortality in 2020; similarly we have also
concerning and likely to deteriorate further in
not accounted for the observed modest reduction in
2021. Returning to pre-pandemic provision is not
trauma related procedures in 2020, which we assume
sufficient, as this will not address the deficit in
will have been treated using alternative strategies, e.g.
joint replacement and even with a rapid expansion
hemiarthroplasty rather than total hip replacement
in service provision to address this deficit in
as was recommended by NHS England in March
provision, our study indicates it will take many
2020, in response to the demands of the pandemic,25
years to resolve this joint replacement crisis.
for patients with intracapsular hip fractures26 or
conservative management.
We are underestimating the impact on elective hip,

shoulder and elbow joint replacement, as surgery
for traumatic indications has been largely preserved
throughout 2020. We also expect a modest lag in data
entry, which principally reduces volume estimates in
the 4th-quarter of 2020 (typically there is less than 5%
late data entry beyond three months). There may also
be a reduction in compliance in reporting procedures
to the NJR because of indirect effects of the allocation
of administrative staff during the COVID-19 pandemic
response in individual hospitals. The model used to
predict time-to-recovery is simplistic and has not
accounted for demographic changes including an
increasingly elderly population nor increasing life
expectancy that we have historically observed. This
analysis only considers the impact of the pandemic
on activity in 2020; the pandemic principally impacted
provision in the last three quarters of 2020 and has
continued to affect provision in 2021 and is likely to
make predictions very conservative.
Declaration of interests • TW reports reimbursement from the National Joint
Registry and honoraria from Smith and Nephew,
• AS, KD, EL, AJ, CLG, EMC, JLR, AP, YBS, MRW, and Pfizer unrelated to this work.
AWB are members of the National Joint Registry
• MRW reports institutional funding from NIHR,
analysis team and contracted to conduct routine
Stryker, Ceramtec, Depuy, Heraeus unrelated
data analysis for the National Joint Registry for
to this work and royalties from Taylor Francis in
quality assurance purposes and routine report
relation to a textbook “Apley & Solomon’s System
generation.
of Orthopaedics and Trauma 10th Edition”.
• AS reports funding from the MRC to his institution
• AWB reports institutional funding from NIHR,
unrelated to this work.
Stryker and Ceramtec unrelated to this work
• EL reports funding from CeramTec to his and royalties from Taylor Francis in relation
institution unrelated to this work. to a textbook “Apley & Solomon’s System of
Orthopaedics and Trauma 10th Edition”.
• JLR reports funding from The National Joint
Registry unrelated to this work. Contributors
• AJ reports personal fees from Freshfields, • AS was responsible for study concept, data
Bruckhaus, Derringer and Anthera analysis, and writing – original draft of the article
Pharmaceuticals Ltd unrelated to this work. – review & editing.
• CLG reports grants from GCRF, ORUK, EBI, • KD and EL were responsible for data analysis,
Wellcome, EDTCP, Versus Arthritis, Chartered writing – review & editing.
Society of Physiotherapists unrelated to this work.
• SKK was responsible for searching literature,
• EMC reports grants from Versus Arthritis and writing – review & editing.
NIHR unrelated to this work.
• AWB, MRW, AJ, YBS, CLG, EMC, JLR, AP,
• MR reports grants from Zimmer, The Health were responsible for study concept, funding
Foundation, Heraeus, 3m Healthcare, Sheulke, acquisition, and writing – review & editing.
Aquilant, Biocomposites, Stryker, Depuy, Smith
and Nephew, Bone Support, NIHR, Ethicon, • MR, TW, MRW, DJP were responsible for study
Convatec unrelated to this work. concept, project administration, and writing –
review & editing.
• DJP reports no conflict of interests.
• AS, KD, EL have accessed and verified the
• JMW reports reimbursement for role as Chairman underlying data.
of the NJR Research Committee and as member
of the NJR Executive Committee. Grant income
from Helmholtz Institute, Munich, Germany;
Amgen, Inc; Versus Arthritis; Wellcome Sanger
Institute; Medical Research Council; Health Quality
Improvement Partnership; NIHR Academic Clinical
Fellowship Programme; NIHR Health Technology
Assessment Programme; NIHR Research Policy
Programme unrelated to this work
• AP reports grant funding from NIHR unrelated to

this work and teaching fees from Zimmer Biomet
unrelated to this work.
References 14. Clement ND, Scott CEH, Murray JRD, Howie CR, Deehan
DJ, Collaboration IM-R. The number of patients “worse
than death” while waiting for a hip or knee arthroplasty
1. Learmonth ID, Young C, Rorabeck C. The operation has nearly doubled during the COVID-19 pandemic. Bone
of the century: total hip replacement. Lancet 2007; Joint J 2021; 103-B(4): 672-80.
370(9597): 1508-19.
15. Jain A, Jain P, Aggarwal S. SARS-CoV-2 Impact on
2. Ben-Shlomo Y, Blom A, Boulton C, et al. The National Elective Orthopaedic Surgery: Implications for Post-
Joint Registry 17th Annual Report 2020. London; 2020. Pandemic Recovery. J Bone Joint Surg Am 2020;
3. Evans JT, Evans JP, Walker RW, Blom AW, Whitehouse 102(13): e68.
MR, Sayers A. How long does a hip replacement last? 16. Brown TS, Bedard NA, Rojas EO, et al. The Effect of
A systematic review and meta-analysis of case series the COVID-19 Pandemic on Electively Scheduled Hip
and national registry reports with more than 15 years of and Knee Arthroplasty Patients in the United States. J
follow-up. Lancet 2019; 393(10172): 647-54. Arthroplasty 2020; 35(7S): S49-S55.
4. Evans JT, Walker RW, Evans JP, Blom AW, Sayers A, 17. Pandit JJ. Pensions, tax and the anaesthetist: significant
Whitehouse MR. How long does a knee replacement last? implications for workforce planning. Anaesthesia 2016;
A systematic review and meta-analysis of case series 71(8): 883-91.
and national registry reports with more than 15 years of
follow-up. Lancet 2019; 393(10172): 655-63. 18. Rimmer A. NHS pensions: what do changes to the lifetime
allowance mean? BMJ 2021; 372: n660.
5. Evans JP, Evans JT, Craig RS, et al. How long does
a shoulder replacement last? A systematic review and 19. Evans JT, Blom AW, Timperley AJ, et al. Factors
meta-analysis of case series and national registry reports associated with implant survival following total hip
with more than 10 years of follow-up. Lancet Rheumatol replacement surgery: A registry study of data from the
2020; 2(9): E539-E48. National Joint Registry of England, Wales, Northern
Ireland and the Isle of Man. PLoS Med 2020; 17(8):
6. Lillie PJ, Samson A, Li A, et al. Novel coronavirus disease e1003291.
(Covid-19): The first two patients in the UK with person to
person transmission. J Infect 2020; 80(5): 578-606. 20. Beard DJ, Davies LJ, Cook JA, et al. The clinical and
cost-effectiveness of total versus partial knee replacement
7. Johnson B. Prime Minister’s statement on coronavirus in patients with medial compartment osteoarthritis
(COVID-19): 23 March 2020. https://www.gov.uk/ (TOPKAT): 5-year outcomes of a randomised controlled
government/speeches/pm-address-to-the-nation-on- trial. Lancet 2019; 394(10200): 746-56.
coronavirus-23-march-2020.
21. Wylde V, Artz N, Marques E, et al. Effectiveness and
8. Collaborative CO. Mortality and pulmonary complications cost-effectiveness of outpatient physiotherapy after
in patients undergoing surgery with perioperative SARS- knee replacement for osteoarthritis: study protocol for a
CoV-2 infection: an international cohort study. Lancet randomised controlled trial. Trials 2016; 17(1): 289.
2020; 396(10243): 27-38.
22. Briggs T. Getting It Right First Time - Orthopaedic
9. Dillon MT, Chan PH, Prentice HA, et al. The effect of a Surgery. 2020. https://www.gettingitrightfirsttime.co.uk/
statewide COVID-19 shelter-in-place order on shoulder surgical-specialty/orthopaedic-surgery/.
arthroplasty for proximal humerus fracture volume and
length of stay. Semin Arthroplasty 2021. 23. Lazic S, Boughton O, Kellett CF, Kader DF, Villet L, Riviere
C. Day-case surgery for total hip and knee replacement:
10. Simon S, Frank BJH, Aichmair A, et al. Impact of the 1st How safe and effective is it? EFORT Open Rev 2018; 3(4):
and 2nd Wave of the COVID-19 Pandemic on Primary 130-5.
or Revision Total Hip and Knee Arthroplasty-A Cross-
Sectional Single Center Study. J Clin Med 2021; 10(6). 24. Judge A, Carr A, Price A, et al. The impact of the
enhanced recovery pathway and other factors on
11. D’Apolito R, Faraldi M, Ottaiano I, Zagra L. Disruption of outcomes and costs following hip and knee replacement:
Arthroplasty Practice in an Orthopedic Center in Northern routine data study. Southampton (UK); 2020.
Italy During the Coronavirus Disease 2019 Pandemic. J
Arthroplasty 2020; 35(7S): S6-S9. 25. England N. Clinical guide for the perioperative care of
people with fragility fractures during the Coronavirus
12. Decruz J, Prabhakar S, Ding BTK, Kunnasegaran R. The pandemic. 2020. https://www.boa.ac.uk/uploads/
COVID-19 pandemic in Singapore: what does it mean for assets/3e2ef4d7-6788-4fcd-8a17196fca00f63e/
arthroplasty? Acta Orthop 2020; 91(5): 551-5. NHSE-Fragility-Fractures-and-Coronavirus-VBOA10-
13. Shih CL, Huang PJ, Huang HT, Chen CH, Lee TC, Hsu 24032020docx.pdf.
CH. Impact of the COVID-19 pandemic and its related 26. National Institute for Health and Care Excellence. Hip
psychological effect on orthopedic surgeries conducted fracture: management. 2017. www.nice.org.uk/guidance/
in different types of hospitals in Taiwan. J Orthop Surg cg124.
(Hong Kong) 2021; 29(1): 2309499021996072.
The effects of the
COVID-19 pandemic on joint
replacement surgery
The patient perspective
By Robin Brittain - Patient Representative, NJR Editorial Board and NJR Steering Committee Member.
Coronavirus (COVID-19) has had a detrimental effect Waiting times for treatment and care have been
on joint replacement surgery waiting times. During an issue throughout the history of the NHS with
the first wave, NHS hospitals were told to suspend all challenges such as increasing demand. Guarantees
non-urgent elective surgery for at least three months on waiting times in several key areas of health care
from 15 April 2020 to help the service deal with the were introduced as part of the Patient Charter in 1991,
COVID-19 pandemic.1 Although elective surgeries which included admission for treatment by a specific
resumed in the UK in mid-2020, most hospitals were date no later than two years from the day someone
subsequently functioning at a much reduced capacity. was placed on a waiting list. By 1995 the guarantee
And with further lockdowns, surgery has been was shortened to a maximum wait for admission to
impacted with further restrictions and reductions in hospital of 18 months.
services and cancellations.
From 2000, progressively tougher targets were
All of this has had an impact on surgical waiting times. introduced. Targets could be seen at times as arbitrary
Other related services have also been affected and and did not always reflect the reality of how long
impacted during this time. Inpatient and critical care people waited. In 2004 a more holistic target was
capacity were freed up and prioritised for COVID-19 introduced, setting an 18-week right under the NHS
patients during the first and subsequent waves. Access Constitution on the total waiting time between GP or
to physiotherapy and occupational therapy which other healthcare professional referral and consultant-
people can be referred to, and which many of them will led treatment in outpatients (non-admitted care) or
benefit from to help with their recovery and aid mobility inpatients (admitted care). This was one of NHS
following surgery, has additionally been affected with England’s significant achievements in the 2000s.2,3,4
reduced services throughout this time.
1 Iacobucci, G. (2020) ‘Covid-19: all non-urgent elective surgery is suspended for at least three months in England’, BMJ (Clinical research ed.), 368, p. m1106.
doi: 10.1136/bmj.m1106.
2 Charlesworth, A., Watt, T. and Gardner, T. (2020) Returning NHS waiting times to 18 weeks for routine treatment. The scale of the challenge pre-COVID-19,
The Health Foundation. Available at: https://www.health.org.uk/publications/long-reads/returning-nhs-waiting-times-to-18-weeks (Accessed: 24 June 2021).
3 Appleby, J. (2010) The waiting game: what’s happening to hospital waiting times?, The King’s Fund. Available at: https://www.kingsfund.org.uk/blog/2010/12/
waiting-game-whats-happening-hospital-waiting-times (Accessed: 24 June 2021).
4 Tudor Edwards, R. (1997) NHS Waiting Lists: Towards The Elusive Solution, Office of Health Economics. Available at: https://www.ohe.org/system/files/private/
publications/228 - 1997_NHS_Waiting_Lists_Edwards.pdf (Accessed: 24 June 2021).
However, despite this change, increasing waiting RTT data show how during the pandemic, trauma and
times for elective joint replacement surgery before the orthopaedic waiting times (including joint replacement
pandemic had already become a reality, not just in surgery), has been steadily increasing across England,
England, but across the UK. There were already issues other than the dip we see below in the Spring and
to meet and satisfy patient needs and expectations Summer of 2020.
with GPs not always referring those for consideration
for joint replacement. This had been despite indicators
to do so, such as pain, and functional limitations and
variations in referral criteria depending on severity of
symptoms and after trying non-surgical treatments
such as physical therapies, weight control and pain
relief.
There were also issues pre-pandemic with increasing

waiting times in accessing certain services, such as
therapy services for rehabilitation to aid recovery post-
surgery. The pandemic exacerbated these issues, for
example, by increasing limitations on getting to see a
GP, let alone requesting them to refer, or for them to Data Source:
be able to refer due to the reduction or suspension of Consultant-led Referral to Treatment Waiting Times Data in
services. England. Incomplete Commissioner Pathway, NHS England.
https://www.england.nhs.uk/statistics/statistical-work-areas/
The COVID-19 virus reached the UK in late January rtt-waiting-times/rtt-data-2020-21
2020, and the waiting times, reduction and recovery
of surgery can be observed subsequent to this time
using different available information. It’s not specifically clear why there is a dip. One
contributor is understood to be a reduction of trauma
Publicly accessible data relating to waiting times referrals, and a factor identified being social distancing
is available and can be useful to assess joint and lockdown measures resulting in less accidents
replacement surgery waiting times but this can be taking place.5,6
limited or have shortcomings.
Finished admission episodes (FAEs), for hip and knee
NHS England collects and publishes monthly joint replacement surgery in England covering a 10-
consultant-led, referral to treatment (RTT) data to year period were released by NHS Digital due to media
monitor meeting the 18-week waiting time target. interest. It covers NHS hospitals and commissioned
This is recorded by clinical specialities. However, a activity in the independent sector. It only relates to hip
drawback is that waiting times cannot be observed and knee joint replacement surgery, but as these are
specifically for joint replacement surgery as this is the most commonly performed implant procedures it’s
included within trauma and orthopaedics (T&O) a useful indicator of waiting times for joint replacement
data. This data is often referenced in regards to surgery as a whole. For the three most recent years,
joint replacement surgery (which does make up a it shows that surgery waiting times have steadily been
reasonable proportion of the data), but it is more of a increasing.
general indicator of waiting time trends.
5 Park, C. et al. (2020) ‘Impact of the COVID-19 pandemic on orthopedic trauma workload in a London level 1 trauma center: the “golden month”’, Acta
Orthopaedica, 91(5), pp. 556–561. doi: 10.1080/17453674.2020.1783621.
6 Sephton, B. M. et al. (2021) ‘The effect of COVID-19 on a Major Trauma Network. An analysis of mechanism of injury pattern, referral load and operative case-
mix’, Injury, 52(3), pp. 395–401. doi: 10.1016/j.injury.2021.02.035.
With those waiting 18 weeks or more: Figures 2 to 4 use NJR acute trauma and elective
data for England, Wales and Northern Ireland, with a
• 2016-2017 = 43,787 people focus on 2020, and with weekly surgical activity and
• 2017-2018 = 45,716 people breakdowns between surgical specialities and also
• 2018-2019 = 55,251 people between nations. This highlights the effects of the first
national lockdown on reductions in volume due to
With those waiting 52 weeks or more: cancelled joint replacement surgery.
• 2016-2017 = 1,320 people It is generally recognised that joint replacement
• 2017-2018 = 1,863 people surgery has one of the longest waiting times7 for
• 2018-2019 = 2,889 people treatment of any speciality. NHS England Referral
to Treatment (RTT) historical data indeed show that
Data Source: trauma and orthopaedics consistently have the longest
Waiting times for hip and knee surgery in England, NHS waiting times, with joint replacement surgery being
Digital. Count of Finished Admission Episodes (FAEs) with a included in this. This is then reflected by how many
main operative procedure of hip or knee replacement with a people are in need of, and receive, such surgery.
treatment waiting times of
The NJR identifies that the primary reason for joint
a) 18 weeks or more, by hospital provider
replacement surgery is arthritis, of which osteoarthritis
b) 52 weeks and more, by hospital provider.
is the most common form of peripheral joint arthritis
Using Hospital Episode Statistics (HES) data.
and cause of disability in the UK.8 The exact incidence
https://digital.nhs.uk/data-and-information/supplementary-
and prevalence of osteoarthritis is difficult to fully
information/2020/waiting-times-for-hip-and-knee-surgery
determine. Around 8.75 million people aged 45 years
and over (33%) in the UK were identified in 2013 by
The National Joint Registry (NJR) collects and records
Versus Arthritis, the leading musculoskeletal support
all joint procedures undertaken, with surgical details,
charity, as seeking treatment for osteoarthritis.9 In the
which can be statistically analysed retrospectively and
absence of any cure, the impact of the national burden
collectively to see surgical activity, patterns and trends,
of osteoarthritis is increasing.
such as those within the timeline of the pandemic.
Therefore, with already rising joint replacement
The section preceding this piece illustrates how the surgery waiting times compounded by the increasing
COVID-19 pandemic has affected surgical activity with incidence of osteoarthritis, COVID-19 has had a
joint replacement procedures undertaken in 2020. further detrimental effect on accessing surgery and on
surgery waiting times.
Figure 1 uses NJR acute trauma and elective data for
England, Wales and Northern Ireland, to show surgical Beyond waiting time data, indicators of joint
activity for a range of past consecutive years, with replacement surgery waiting times can include the use
procedures undertaken steadily increasing until 2020 of anecdotal information, including the testimonials
when they decline with the start of the pandemic and, of those waiting for surgery, survey work, and data
as previously mentioned, waiting times also increase analysis, which can be part of research and reporting
leading up to 2020, increasing more with the effects of work by charities, patient groups, institutions and
the pandemic.
7 National Health Service England (no date) Consultant-led Referral to Treatment Waiting Times. Available at: https://www.england.nhs.uk/statistics/statistical-
work-areas/rtt-waiting-times (Accessed: 16 June 2021).
8 Swain, S. et al. (2020) ‘Trends in incidence and prevalence of osteoarthritis in the United Kingdom: findings from the Clinical Practice Research Datalink
(CPRD)’, Osteoarthritis and Cartilage, 28(6), pp. 792–801. doi: 10.1016/j.joca.2020.03.004.
9 Versus Arthritis (2019) The State of Musculoskeletal Health 2019. Arthritis and other musculoskeletal conditions in numbers. Available at: https://www.
versusarthritis.org/media/14594/state-of-musculoskeletal-health-2019.pdf.
academics.10 Additionally, modelling of the various for people waiting for surgery.
sources of data can take place, such as by Mishra et
al.11 and Oussedik et al.12 “Physically on a day-to-day basis it’s really
changed my life completely. I can’t just get up
It can be just as important to make use of a variety and walk out of the house and go wherever I
of resources and not just pure statistical data to want to go. I have to live with a pain threshold
understand the waiting time landscape. And in and it’s there, it affects my walking, my sitting,
particular, the personal stories of those waiting for my standing. My sleeping is impacted by it,
surgery can highlight the physical, mental, emotional I don’t have a full night’s sleep anymore. It’s
and occupational and social effects that waiting for horrible, it’s just a nagging pain. No medication
joint replacement surgery has on them, and which has helped me. It’s basically only surgery that
data doesn’t typically show or reflect. will fix me now.”
- Rob Martinez, June 2020.
Rob was put on the waiting list in October 2019

The resulting impact for double knee joint replacement surgery and was
on patients due to have his first knee replaced in April 2020, but
was notified in March that his operation had been
People in need of joint replacement surgery report cancelled. He had his right knee replaced in October
symptoms of pain, even when at rest, alongside aches 2020.13,14
following activity, and limited or loss of function which
There can be increased resulting disability due to
include joint stiffness, limited range of movement
worsening function, decreased mobility and pain.
and mobility, difficulty with sleep due to the pain and
associated tiredness and fatigue. The risk related to delaying joint replacement surgery
is that it may lead to the deterioration of the joint,
There can be knock-on effects of increased difficulty
with damage inside and outside of the joint, or even
with undertaking and carrying out tasks and activities
deformity. Additionally, there can be risks to muscles,
with normal daily living, socially and with work,
ligaments and other structures becoming weak and
all which can have a resulting financial toll and
losing function and strength, for example with muscle
implications. There are those who because of their
wasting and deficit. Prolonged delays may result in
poor health-state can simply have difficulty even
reduced surgical options and joint replacement may
leaving their homes.
become a more complicated process with possible
This all has a negative effect on their quality of life. longer surgery than is normal and with increased
anaesthesia use. All of this has implications with
Timely access to joint replacement surgery is various risk and effects, including the resulting impact
important to ensure that people can benefit from the on recovery time.
easing of symptoms, leading to a better quality of
life. COVID-19 has exacerbated the severity of the With delays, there can be an inability for a patient to
situation with increasing issues and painful symptoms manage and cope with pain, and with limited pain
management options being offered or available, there
10 Versus Arthritis (2021) Supporting People with Arthritis Waiting for Surgery. A six-part package to support people with arthritis waiting for joint replacement
surgery in England. Available at: https://www.versusarthritis.org/media/23694/joint-replacement-support-package-june2021.pdf.
11 Mishra, B., BODS Collaborators and Roy, B. (2020) ‘BODS/BOA Survey of impact of COVID-19 on UK orthopaedic practice and implications on restoration of
elective services - Part 2’, The Transient Journal.
12 Oussedik, S. et al. (2021) ‘Elective orthopaedic cancellations due to the COVID-19 pandemic: where are we now, and where are we heading?’, Bone & Joint
Open, 2(2), pp. 103–110. doi: 10.1302/2633-1462.22.bjo-2020-0161.r1.
13 Life ‘on hold’ for Berkshire man waiting for knee replacement (2020) ITV News. Available at: https://www.itv.com/news/meridian/2020-06-19/life-on-hold-for-
berkshire-man-waiting-for-knee-replacement (Accessed: 21 May 2021).
14 Chalmers, V. (2021) UNITED IN PAIN Feeling suicidal, plagued by stabbing pains – we are the faces behind record NHS waiting lists, The Sun. Available at:
https://www.thesun.co.uk/fabulous/14942694/faces-behind-record-nhs-waiting-lists-suicidal-pain-arthritis (Accessed: 26 May 2021).
can be a reliance on the use of strong pain relief into my hip, groin and then that just completely
medication such as opioids. This brings with it the knocks your mental health.”
potential risks and effects of experiencing drowsiness,
clouded thinking or ‘brain fog’, and anxiety amongst “I just felt that I couldn’t go on any more. I didn’t
other possible issues, as well as the potential risk of want to be here any more. What was the point
over-prescribing and/or misuse which can lead to because all in front of me all I could see was this
adverse events. waiting list going on and on and on.”
On top of all of this, there have also been COVID-19 - Liz McLucas, 25th February 2021.
lockdown rules to adhere to and cope with, and for Liz was having to take painkillers, including morphine,
those identified as clinically extremely vulnerable due to control the pain.
to having certain health conditions, undergoing certain
treatments and taking certain drug medication which “No matter how much pain killers they gave me it
can suppress immune systems, there has wasn’t enough.”
been the added burden of following advice for
“It took the edge of it but it took the edge off
protective shielding.
everything else”.
If one is less active and more sedentary, particularly
“It got to the stage where getting out of bed was
over a long period of time, there is the potential risk of
nearly too much.”
other health issues, co-morbidities and complications
developing such as weight gain leading to obesity, Liz took the decision to pay for private surgery, despite
type 2 diabetes, cardiovascular and respiratory being against private healthcare in principle, due to
disorders, which in turn might have implications, such being on a waiting list for over two years. It cost her
as risk for undergoing joint replacement surgery. £10,800, borrowing money from her two sons to pay
for it. She had hip replacement surgery in January
The waiting, the not knowing when surgery might
2021, after chronic pain had confined her to bed for
happen, and feelings of being abandoned, isolated
much of the previous 12 months.
and trapped, combined with dealing with pain, limited
and even loss of function and mobility, sleep issues, “What mother wants to turn round and ask her
tiredness and fatigue, mood swings with frustration, children for money?”15
irritability and anger, and ultimately reduced quality
of life can additionally have an effect on mental A multi-centre cross-sectional study in 2020 by Scott
health and wellbeing resulting in worry, anxiety and et al. concluded that one-third of patients waiting for
depression. total hip arthroplasty and nearly one-quarter waiting
for a knee arthroplasty procedure were categorised in
“If I hadn’t had that operation I don’t think I’d be a state “worse than death” (patients scoring less than
talking to you today because I could not have zero for their EQ-5D score). And that every increasing
carried on the way things were…” six-month period a patient waited for surgery was
associated with a clinically significant deterioration in
“My whole life was just basically gone. I had no
the quality of their life.16
life left.”
“My mobility is nearly zero as a result of the
“The pain was just getting more and more
excessive pain. It never stops; it is just constant
excruciating. I couldn’t even stand up - it got to
... I’ve begun to feel like life is not worth living.”
the stage where I couldn’t put weight on it.”
- Christopher Bulteel, age 72, October 2020.
“It was like somebody was just constantly sawing
15 McKeown, L.-A. (2021) Liz McLucas: ‘I had to borrow money from my sons for surgery’, BBC News. Available at: https://www.bbc.co.uk/news/uk-northern-
ireland-56195209 (Accessed: 21 May 2021).
16 Clement, N. D. et al. (2021) ‘The number of patients “worse than death” while waiting for a hip or knee arthroplasty has nearly doubled during the COVID-19
pandemic’, The bone & joint journal, 103-B(4), pp.672–680. doi: 10.1302/0301-620X.103B.BJJ-2021-0104.R1.
Christopher was due to have a hip replacement in receiving surgery was already taking place long before
March 2020 after being on a waiting list for a year, but the arrival of the COVID-19 pandemic.
the operation was cancelled due to the coronavirus
outbreak, leaving him virtually housebound.17,18 The Versus Arthritis launched their ‘Right on Time’
NJR was in contact with Christopher in early August campaign in February 2020 calling for improved
2021 and he said that when he had been feeling low, access to joint replacement surgery due to increasing
thinking about his family and others worse off than waiting times. The campaign was paused with the
himself had helped him. He had still not had surgery. arrival of the pandemic. They then launched their
‘Impossible to Ignore’ campaign in July 2020 to
When surgery has been able to take place, for some ensure the wider needs of those with arthritis are
there have been concerns, worry and being anxious recognised and addressed by the government and
and apprehensive about not only being safe going policy makers so that they are not left behind, for
into hospital, but also while having surgery and being example being in pain, and that people with arthritis
cared for, with the risk of contracting COVID-19. A and related conditions:
study of 102 patients who were on the waiting list
of a single high-volume procedure surgeon, having • can access healthcare services throughout the
previously been given a date for surgery for an pandemic;
elective hip or knee procedure during the COVID-19 • know that there will be a commitment to their
pandemic, identified the number of patients wanting ongoing involvement in shaping the future of services
to proceed with their planned elective surgery in the and treatment;
COVID-19 environment. Overall, 58 patients (56.8%) • have timely and clear communication and access to
preferred to continue with planned surgery upon advice and support to manage their pain.
resumption of elective orthopaedic services, in spite of
additional risks posed by COVID-19. This leaves nearly Also, with regards to joint replacement surgery, that
half who were not willing or wanting to have surgery.19 it is able to continue wherever it is safe to do so
throughout the pandemic, and that there is a national
Patient groups, charities and representative bodies plan with action to bring down joint replacement
including Versus Arthritis and the Arthritis and waiting lists.
Musculoskeletal Alliance (ARMA), the umbrella body
for organisations providing musculoskeletal services, With the resumption of elective surgery there is a
have for a number of years been calling for improved backlog of procedures that need to be undertaken.
access to joint replacement surgery. The NJR has an important role to play. There is an
even greater emphasis and focus for the registry to
Campaigns have addressed not only long waiting monitor joint replacement surgery to ensure that the
lists but also the restrictions and rationing that highest quality of outcome standards in terms of
has been taking place, for example, overweight or surgeon and implant performance, alongside patient
obese patients being told to lose weight or others outcomes and safety that have come to be expected,
to stop smoking in order to be a suitable candidate continue despite the volume and hospital pressures
for surgery, as opposed to clinical need.20,21,22 And that will co-exist to reduce these waiting list times.
the practice of what can been seen as a barrier to
17 Tanner, C. (2020) ‘I’ve begun to feel like life is not worth living’ says man, 71, who has waited 21 months for hip replacement, inews.co.uk. Available at: https://
inews.co.uk/news/real-life/waiting-lists-nhs-figures-routine-operations-arthritis-knee-hip-replacements-790201 (Accessed: 21 May 2021).
18 Durkin, J. (2021) Former mayor living in constant pain after op cancelled, Bournemouth Echo. Available at: https://www.bournemouthecho.co.uk/
news/19047434.former-mayor-living-constant-pain-op-cancelled (Accessed: 21 May 2021).
19 Chang, J. et al. (2020) ‘Restarting elective orthopaedic services during the COVID-19 pandemic. Do patients want to have surgery?’, Bone & Joint Open, 1(6),
pp. 267–271. doi: 10.1302/2633-1462.16.bjo-2020-0057.
20 Arthritis and Musculoskeletal Alliance (ARMA) (2017) Policy Position Paper. ‘Rationing’ Access to Joint Replacement Surgery and Impact on People with Arthritis
and Musculoskeletal Condition. Available at: http://arma.uk.net/wp-content/uploads/2017/08/Policy-Position-Paper-Surgery_v5_Interactive.pdf.
21 The Association of British HealthTech Industries (ABHI) (2017) Hip and Knee Replacement: The Hidden Barriers. Available at: https://www.abhi.org.uk/
media/1379/hip-and-knee-replacement-the-hidden-barriers.pdf.
22 Dodd, L. et al. (2015) ‘Rationing of orthopaedic surgery in the UK’, Bone & Joint 360, 4(6), pp. 2–5. doi: 10.1302/2048-0105.45.360391.
Glossary
ABHI Association of British HealthTech Industries – the UK trade association of medical device suppliers.
Acetabular component The portion of a total hip replacement prosthesis that is inserted into the acetabulum – the socket part
of a ball and socket joint.
Acetabular cup See Acetabular component.
Acetabular prosthesis See Acetabular component.
Administrative censoring Administrative censoring is the process of defining the end of the observation period for the cohort. All
patients are assumed to have experienced either a revision, be dead or alive at the censoring date.
ALVAL Aseptic Lymphocyte-dominated Vasculitis-Associated Lesion. This term is used in the Annual Report
to describe the generality of adverse responses to metal debris, but in its strict sense refers to the
delayed type-IV hypersensitivity response.
Amputation The surgical removal of a limb.
Antibiotic-loaded bone cement A bone cement which contains pre-mixed antibiotics, this is distinct from plain bone cement which
contains no antibiotics. See Bone cement.
Arthrodesis A procedure where the bones of a natural joint are fused together (stiffened).
Arthroplasty A procedure where a native joint is surgically reconstructed or replaced with an artificial prosthesis.
ASA American Society of Anesthesiologists scoring system for grading the overall physical condition of
the patient, as follows: P1 – fit and healthy; P2 – mild disease, not incapacitating; P3 – incapacitating
systemic disease; P4 – life threatening disease; P5 – expected to die within 24 hrs without an operation.
BASK British Association for Surgery of the Knee.

Bearing type The two surfaces that articulate together in a joint replacement. Options described in the report
include metal-on-polyethylene, metal-on-metal, ceramic-on-polyethylene, ceramic-on-metal, ceramic-
on-ceramic and in dual mobility hip replacements metal-on-polyethylene-on-metal and ceramic-on-
polyethylene-on-metal.
BESS British Elbow and Shoulder Society.
Beyond Compliance A system of post market surveillance initiated in 2013. Under this system, Beyond Compliance
collates NJR data, national PROMs and data from implanting surgeons, and monitors the usage and
performance of implants which are new to the market.
BHS British Hip Society.
Operation performed on both sides, e.g. left and right knee procedures, carried out on the same day
Bilateral operation
or on different days.
BOA British Orthopaedic Association. The surgical specialty association for trauma and orthopaedics in
the UK.
Body mass index (BMI) A statistical tool used to estimate a healthy body weight based on an individual’s height. The BMI is
calculated by dividing a person’s weight (kg) by the square of their height (m2).
BOFAS British Orthopaedic Foot and Ankle Society.
Bone cement The material used to fix cemented joint replacements to bone – polymethyl methacrylate (PMMA).
Brand (of prosthesis) The brand of a prosthesis (or implant) is the manufacturer’s product name, e.g. the Exeter V40 brand
for hips, the PFC Sigma brand for knees, the Zenith brand for ankles, the Delta Xtend brand for
shoulders and the Coonrad Morrey for elbows.
C
Case ascertainment Proportion of all relevant joint replacement procedures performed that are entered into the NJR.
Case mix Term used to describe variation in surgical practice, relating to factors such as indications for surgery,
patient age and gender.
Cement See Bone cement.
Cemented Prostheses designed to be fixed into the bone using bone cement.
Cementless See Uncemented.
Compliance The percentage of total joint procedures that have been entered into the NJR where the denominator
is defined as the number of all eligible procedures.
Confidence Interval (CI) A ‘Confidence Interval’ (CI) illustrates the uncertainty of an estimated statistic. For example, a CI for the
cumulative probability of revision tells us the probability that ‘true’ (population) probability of revision
will fall between the range of values on a specified percentage, typically 95%, of occasions if the data
collection was repeated.
Confounding Confounding occurs when either a measured or unmeasured factor (variable) distorts the true
relationship between the exposure and outcome of interest. For example, a comparison of the revision
rates between two distinct types of implant may be 'confounded' because one implant has been used
on an older group of patients compared to the other. In this context, age may be a 'confounder' if it
distorts the relationship between implant type and outcome i.e. revision rate. Statistical methods may
help to ‘adjust’ for such confounding factors however residual confounding of an association may
always persist.
Conventional total shoulder Replacement of the shoulder joint which replicates the normal anatomical features of a shoulder joint.
replacement
Coverage Scope of inclusion criteria for the registry. Data submission has been mandatory for independent
organisations since 1 April 2003 and for NHS organisations since 1 April 2011. See also NJR definition.
COVID-19 Coronavirus disease following infection from the SARS-CoV-2 virus.
Cox ‘proportional hazards’ model A type of multivariable regression model used in survival analysis to look at the effects of a number of
variables (‘exposures’) on outcome (first revision or death). The effect of each variable is adjusted for
the effects of all the other ‘exposure’ variables in the model. Some regression models used in survival
modelling make assumptions about the way the hazard rate changes with time (see ‘hazard rate’).
The Cox model doesn’t make any assumptions about how the hazard rate changes, however it does
assume that the exposure variables affect the hazard rates in a ‘proportional’ way.
CQC Care Quality Commission. Regulators of care provided by the NHS, local authorities, private
companies and voluntary organisations.
Cumulative Incidence Function A different way of estimating failure compared to Kaplan-Meier, see Kaplan-Meier. Also known as
(CIF) observed or crude failure, as the estimate reflects what is seen in practice.
Cup See Acetabular component.
DAIR Debridement And Implant Retention. In cases of infection, the surgeon may debride (surgically clean)
the surgical site and retain the joint replacement implants. The NJR does not collect data on Antibiotic
use and therefore DAIR in our context focuses on implant and procedure data.
DAIR with Modular Exchange Debridement And Implant Retention with Modular Exchange. In cases of infection where the implants
are modular, the surgeon may debride (surgically clean) the surgical site, exchange the modular
components (e.g. head, acetabular liner) and retain the non-modular joint replacement implants.
Data collection periods for annual Outcomes analyses present data for hip, knee, ankle, elbow and shoulder procedures that took
report analysis place between 1 April 2003 and 31 December 2020 inclusive. Hospital (unit) level analyses present
data for hip and knee procedures undertaken between 1 January and 31 December 2020 inclusive.
Online interactive reporting presents data for each calendar year - 1 January to 31 December
inclusive. Hospital (unit) outlier analysis is performed on the last five and ten years of data up to 14
February 2021.
DDH Developmental dysplasia of the hip. A condition where the hip joint is malformed, usually with a shallow
socket (acetabulum), which may cause instability.
Distal humeral hemiarthroplasty A type of elbow replacement which only replaces the distal part of the humerus.
DHSC Department of Health and Social Care.
Dual mobility Dual mobility is a type of total hip replacement which contains two articulating bearing surfaces. The
distal bearing surface consists of a standard femoral head which articulates within a large polyethylene
bearing. The proximal bearing surface consists of an acetabular bearing which articulates against
a large polyethylene bearing. The femoral head and acetabular bearing can be made of metal or
ceramic.
DVT Deep vein thrombosis. A blood clot that can form in the veins of the leg and is recognised as a
significant risk after joint replacement surgery.
Episode An event involving a patient procedure such as a primary or revision total prosthetic replacement.
An episode can also consist of two consecutive procedures, e.g. a stage one of two-stage revision,
followed by a stage two of two-stage revision.
Excision arthroplasty A procedure where the articular ends of the bones are simply excised, so that a gap is created
between them, or when a joint replacement is removed and not replaced by another prosthesis.
Femoral component (hip) Part of a total hip joint that is inserted into the femur (thigh bone) of the patient. It normally consists of a
stem and head (ball).
Femoral component (knee) Portion of a knee prosthesis that is used to replace the articulating surface of the femur (thigh bone).
Femoral head Spherical portion of the femoral component of the artificial hip replacement. May be modular or non-
modular i.e. attached to the stem, see monobloc.
Femoral prosthesis Portion of a total joint replacement used to replace damaged parts of the femur (thigh bone).
Femoral stem The part of a modular femoral component inserted into the femur (thigh bone). It has a femoral head
mounted on it to form the complete femoral component in hip replacement or may be added to the
femoral component of a total knee replacement, usually in the revision setting.
Funnel plot A graphical device to compare unit or surgeon performance. Measures of performance (e.g. a ratio
of number of observed events to the expected number based on case-mix) are plotted against an
interpretable measure of precision. Control limits are shown to indicate acceptable performance. Points
outside of the control limits suggest ‘special cause’ as opposed to ‘common cause’ variation (see for
example D Spiegelhalter, Stats in Medicine, 2005).
Glenoid component The portion of a total shoulder replacement prosthesis that is inserted into the scapula – the socket
part of a ball and socket joint in conventional shoulder replacement or the ball part in reverse
shoulder replacement.
H
Hazard rate Rate at which ‘failures’ occur at a given point in time after the operation conditional on ‘survival’ up
to that point. In the case of first revision, for example, this is the rate at which new revisions occur in
those previously unrevised.
Head See Femoral head and/or Humeral head and/or Radial head component (elbow).
Healthcare provider NHS or independent sector organisation that provides healthcare; in the case of the NJR, orthopaedic
hip, knee, ankle, elbow or shoulder replacement surgery.
HES Hospital Episode Statistics. A data source managed by NHS Digital which contains data on conditions
(ICD-10 codes), procedures (OPCS-4 codes) in addition to other hospital statistics collected routinely
by NHS hospitals in England.
Highly cross-linked polyethylene See Modified Polyethylene.
HQIP Healthcare Quality Improvement Partnership. Hosts the NJR on behalf of NHS England.
Promotes quality in health and social care services and works to increase the impact that clinical
audit has nationally.
Humeral component (elbow/distal) Part of a total elbow joint that is inserted into the humerus (upper arm bone) of the patient to replace
the articulating surface of the humerus.
Humeral component (shoulder/ Part of a total or partial shoulder replacement that is inserted into the humerus (upper arm bone) of the
proximal) patient. It normally consists of a humeral stem and head (ball) in conventional shoulder replacement or
a humeral stem and a humeral cup in a reverse shoulder replacement.
Humeral head Domed head portion of the humeral component of the artificial shoulder replacement attached to the
humeral stem.
Humeral prosthesis Portion of a shoulder replacement used to replace damaged parts of the humerus (upper arm bone).
Humeral stem The part of a modular humeral component inserted into the humerus (upper arm bone). Has a humeral
head or humeral cup mounted on it to form the complete humeral implant.
Hybrid procedure Joint replacement procedure in which cement is used to fix one prosthetic component while the other
is cementless. For hip procedures, the term hybrid covers both reverse hybrid (uncemented stem,
cemented socket) and hybrid (cemented stem, uncemented socket) unless separately defined.
ID A generic term for pseudo anonymised patient identification number, whether that be a pseudo
anonymised NHS number, local hospital patient identifier or combination of personal characteristics.
Image/computer-guided surgery Surgery performed by the surgeon, using real-time images and data computed from these to assist
alignment and positioning of prosthetic components.
Inconsistent operative pattern A sequence of operations where the primary operation is not the first operation in the sequence or
where there are multiple primary operations.
Independent hospital A hospital managed by a commercial company that predominantly treats privately-funded patients but
does also treat NHS-funded patients.
Index joint The primary joint replacement that is the subject of an NJR entry.
Indication (for surgery) The reason for surgery. The NJR system allows for more than one indication to be recorded.
Ipsilateral procedure An operation performed on one side, e.g. left or right knee procedures.
IQR The interquartile range shows a range of values from the 25th (first quartile) and 75th (third quartile)
centiles of a variables distribution.
ISTC Independent sector treatment centre. See Treatment centre.
Kaplan-Meier Used to estimate the cumulative probability of ‘failure’ at various times from the primary operation, also
known as Net Failure. ‘Failure’ may be either a first revision or a death, depending on the context. The
method properly takes into account ‘censored’ data. Censorings arise from incomplete follow-up; for
revision, for example, a patient may have died or reached the end of the analysis period (end of 2020)
without having been revised.
Lateral resurfacing (elbow) Partial resurfacing of the elbow with a humeral surface replacement component used with a lateral
resurfacing head inserted with or without cement.
LHMoM Large head metal-on-metal. Where a metal femoral head of 36mm diameter or greater is used in
conjunction with a femoral stem, and is articulating with either a metal resurfacing cup or a metal liner
in a modular acetabular cup. Resurfacing hip replacements are excluded from this group.
Linkable percentage Linkable percentage is the percentage of all relevant procedures that have been entered into the NJR,
which may be linked via NHS number to other procedures performed on the same patient.
Linkable procedures Procedures entered into the NJR database that are linkable to a patient’s previous or subsequent
procedures by the patient’s NHS number.
Linked total elbow Where the humeral and ulnar parts of a total elbow replacement are structurally coupled.
LMWH Low molecular weight Heparin. A blood-thinning drug used in the prevention and treatment of deep
vein thrombosis (DVT).
MDS Minimum dataset, the set of data fields collected by the NJR. Some of the data fields are mandatory
(i.e. they must be filled in). Fields that relate to patients’ personal details must only be completed where
informed patient consent has been obtained.
MDSv1 Minimum dataset version one, used to collect data from 1 April 2003. MDSv1 closed to new data entry
on 1 April 2005.
MDSv2 Minimum dataset version two, introduced on 1 April 2004. MDSv2 replaced MDSv1 as the official
dataset on 1 June 2004.
MDSv3 Minimum dataset version three, introduced on 1 November 2007 replacing MDSv2 as the new official
dataset.
MDSv4 Minimum dataset version four, introduced on 1 April 2010 replacing MDSv3 as the new official dataset.
This dataset has the same hip and knee MDSv3 dataset but includes the data collection for total ankle
replacement procedures.
MDSv5 Minimum dataset version five, introduced on 1 April 2012 replacing MDSv4 as the new official dataset.
This dataset has the same hip, knee and ankle MDSv4 dataset but includes the data collection for total
elbow and total shoulder replacement procedures.
MDSv6 Minimum dataset version six, introduced on 14 November 2014 replacing MDSv5 as the new official
dataset. This dataset includes the data collection for hip, knee, ankle, elbow and shoulder replacement
procedures.
MDSv7 Minimum dataset version seven, introduced on 4 June 2018 replacing MDSv6 as the new official
dataset. This dataset includes reclassification and amendments to data collection for hip, knee, ankle,
elbow and shoulder replacement procedures.
MHRA Medicines and Healthcare products Regulatory Agency. The UK regulatory body for medical devices.
Minimally-invasive surgery Surgery performed using small incisions (usually less than 10cm). This may require the use of special
instruments.
Mix and match Mix and match describes when the components of the joint construct come from different brands and/
or manufacturers.
Modified Polyethylene (MP) Any component made of polyethylene which has been modified in some way in order to improve its
performance characteristics. Some of these processes involve chemical changes, such as increasing
the cross-linking of the polymer chains or the addition of vitamin E and/or other antioxidants. Others
are physical processes such as heat pressing or irradiation in a vacuum or inert gas.
Modular Component composed of more than one piece, e.g. a modular acetabular cup shell component with a
modular cup liner, or femoral stem coupled with a femoral head.
Monobloc Component composed of, or supplied as, one piece, the antonym of modular e.g. a monobloc knee
tibial component.
Multicompartmental knee More than one compartmental knee replacement within the same operation e.g. a unicondylar
replacement knee replacement and patellofemoral knee replacement, a medial and a lateral unicondylar knee
replacement or a medial and a lateral and patellofemoral unicondylar knee replacement.
NHS National Health Service (E – England, I – Improvement, X – Digital).

NHS No. Pseudo anonymised National Health Service Number.
NICE National Institute for Health and Care Excellence.
NICE benchmark The NICE benchmark of performance is defined as a 5% prosthesis failure rate at 10 years.
NJR The National Joint Registry (NJR), which covers England, Wales, Northern Ireland, the Isle of Man and
the States of Guernsey, has collected and analysed information from both the NHS and independent
healthcare sectors on hip and knee replacements since 1 April 2003, ankle replacements since 1 April
2010, and elbow and shoulder replacements since April 2012.
NJR Centre National coordinating centre for the NJR.
NJR Stats Online Online facility for viewing and downloading NJR statistics on www.njrcentre.org.uk/njrcentre/
Healthcare-providers/Accessing-the-data/StatsOnline/NJR-StatsOnline.
ODEP Orthopaedic Data Evaluation Panel of the NHS Supply Chain. www.odep.org.uk.
ODEP ratings A letter and star rating awarded to implants based on their performance at specified time points. See
www.odep.org.uk for more details.
OPCS-4 Office of Population, Censuses and Surveys: Classification of Interventions and Procedures, version 4
– a list of surgical procedures and codes.
Outlier Data for a surgeon, unit or implant brand that falls outside of acceptable control limits. See also ‘Funnel
plot’. A Level One implant outlier is defined as having a PTIR of more than twice the group average. A
Level Two implant outlier is defined as having a PTIR of 1.5 times the group average.
Patellar resurfacing Replacement of the surface of the patella (knee cap) with a prosthesis.
Patellofemoral knee replacement Procedure involving replacement of the trochlear and replacement resurfacing of the patella.
Patellofemoral prosthesis Two-piece knee prosthesis that provides a prosthetic (knee) articulation surface between the patella
and trochlear.
Patient consent Patient personal details may only be submitted to the NJR where explicit informed patient consent has
been given or where patient consent has not been recorded. If a patient declines to give consent, only
the anonymous operation and implant data may be submitted.
Patient physical status See ASA.

PDS The NHS Personal Demographics Service is the national electronic database of NHS patient
demographic details. The NJR uses the PDS Demographics Batch Service (DBS) to source missing
NHS numbers and to determine when patients recorded on the NJR have died.
PEDW Patient Episode Database for Wales. The Welsh equivalent to Hospital Episode Statistics (HES) in
England.
Primary hip/knee/ankle/elbow/ The first time a joint replacement operation is performed on any individual joint in a patient.
shoulder replacement
Procedure A single operation. See also Primary hip/knee/ankle/elbow/shoulder replacement and Revision hip/
knee/ankle/elbow/shoulder replacement.
PROM(s) Patient Reported Outcome Measure(s).
Prosthesis Orthopaedic implant used in joint replacement procedures, e.g. a total hip, a unicondylar knee, a total
ankle, a reverse shoulder or a radial head replacement.
Prosthesis-time The total of the length of time a prosthesis was ‘at risk’ of revision. In the calculation of PTIRs for
revision, for example, each individual prosthesis construct time is measured from the date of the
primary operation to the date of first revision or, if there has been no revision, the date of patient’s
death or the administrative censoring date.
Proximal humeral hemiarthroplasty A shoulder replacement procedure which replaces only the humeral side of the shoulder joint.
PTIR Prosthesis-Time Incidence Rate. The total number of events (e.g. first revisions) divided by the total of
the lengths of times the prosthesis was at risk (see ‘Prosthesis-time’).
Pulmonary embolism A pulmonary embolism is a blockage in the pulmonary artery, which is the blood vessel that carries
blood from the heart to the lungs.
Radial head component (elbow) Part of a partial elbow joint that is inserted into the radius (outer lower arm bone) of the patient to
replace the articulating surface of the radial head. May be monobloc or modular.
Region NJR regions are based on the former NHS Strategic Health Authority areas. These organisations were
responsible for managing local performance and implementing national policy at a regional level until
2013.
Resurfacing (hip) Resurfacing of the femoral head with a surface replacement femoral prosthesis and insertion of a
monobloc acetabular cup, with or without cement.
Resurfacing (knee) See Patellar resurfacing.
Resurfacing (shoulder) Resurfacing of the humeral head with a surface replacement humeral prosthesis inserted, with or
without cement.
Reverse polarity total shoulder Replacement of the shoulder joint where a glenoid head is attached to the scapula and the humeral
replacement cup to the humerus.
Revision burden The proportion of revision procedures carried out as a percentage of the total number of surgeries on
that particular joint.
Revision hip/knee/ankle/elbow/ A revision is defined as any operation where one or more components are added to, removed from
shoulder replacement or modified in a joint replacement or if a Debridement And Implant Retention (DAIR) with or without
modular exchange is performed. Capturing DAIR with or without modular exchange commenced with
the introduction of MDSv7. Prior to this DAIR with modular exchange was included as a single-stage
revision but DAIR without modular exchange was not captured. Within the annual report, each of these
procedure types is included in the analyses as a revision episode. This is distinct from the analyses in
the surgeon, unit, and implant performance work streams where DAIR without modular exchange is
not currently included as a revision outcome.
S
Shoulder humeral hemiarthroplasty Replacement of the humeral head with a humeral stem and head or shoulder resurfacing component
which articulates with the natural glenoid.
Single-stage revision A complete revision procedure carried out in a single operation, i.e. components removed and
replaced under one anaesthetic.
SOAL Lower Layer Super Output Areas. Geographical areas for the collection and publication of small area
statistics. These are designed to contain a minimum population of 1,000 and a mean population size
of 1,500. Please also see Office for National Statistics at www.ons.gov.uk.
Stemless shoulder replacement A shoulder replacement where the most distal element of humeral section does not project beyond the
metaphyseal bone of the proximal humerus.
Stemmed shoulder replacement A shoulder replacement where the most distal element of humeral section projects into the diaphysis of
the proximal humerus.
Subtalar The joints between the talus and the calcaneum, also known as the talocalcaneal joints.
Surgical approach Method used by a surgeon to gain access to, and expose, the joint.
Survival (or failure) analysis Statistical methods to look at time to a defined failure ‘event’ (for example either first revision or death);
see Kaplan-Meier estimates and Cox ‘proportional hazards’ models. These methods can take into
account cases with incomplete follow-up (‘censored’ observations).
Talar component Portion of an ankle prosthesis that is used to replace the articulating surface of the talus at the ankle
joint.
TAR Total ankle replacement (total ankle arthroplasty). Replacement of both tibial and talar surfaces, in most
cases implanted without cement.
TED stockings Thrombo embolic deterrent (TED) stockings. Elasticised stockings that can be worn by patients
following surgery and which may help reduce the risk of deep vein thrombosis (DVT).
THR Total hip replacement (total hip arthroplasty). Replacement of the femoral head with a stemmed femoral
prosthesis and insertion of an acetabular cup, with or without cement.
Thromboprophylaxis Drug or other post-operative regime prescribed to patients with the aim of preventing blood clot
formation, usually deep vein thrombosis (DVT), in the post-operative period.
Tibial component (ankle) Portion of an ankle prosthesis that is used to replace the articulating surface of the tibia (shin bone) at
the ankle joint.
Tibial component (knee) Portion of a knee prosthesis that is used to replace the articulating surface of the tibia (shin bone) at
the knee joint. May be modular or monobloc (one piece).
TKR Total knee replacement (total knee arthroplasty). Replacement of both tibial and femoral condyles (with
or without resurfacing of the patella), with or without cement.
Total condylar knee Type of knee prosthesis that replaces the complete contact area between the femur and the tibia of a
patient’s knee.
Total elbow replacement Replacement of the elbow joint which consists of both humeral and ulna prostheses.
Treatment centre Treatment centres are dedicated units that offer elective and short-stay surgery and diagnostic
procedures in specialties such as ophthalmology, orthopaedic and other conditions. These include
hip, knee, ankle, elbow, and shoulder replacements. Treatment centres may be privately funded
(independent sector treatment centre – ISTC). NHS Treatment Centres exist but their data is included
in those of the English NHS Trusts and Welsh Local Health Boards to which they are attached.
Trochanter Bony protuberance of the femur, the greater trochanter is found on its upper outer aspect and is the
site of attachment of the abductor muscles. The lesser trochanter is medial and inferior to this and is
the site of attachment of the psoas tendon.
Trochanteric osteotomy A procedure to temporarily remove and then reattach the greater trochanter, used to aid exposure of
hip joint during some types of total hip replacement and now usually used only in complex procedures.
Two-stage revision A revision procedure carried out as two operations, i.e. under two separate anaesthetics, most often
used in the treatment of prosthetic joint infection.
Type (of prosthesis) Type of prosthesis is the generic description of a prosthesis, e.g. modular cemented stem (hip),
patellofemoral joint (knee), talar component (ankle), reverse shoulder (shoulder) and radial head
replacement (elbow).
Ulnar component (elbow) Part of a total elbow joint that is inserted into the ulna (inner lower arm bone) of the patient to replace
the articulating surface of the ulna. May be linked or unlinked.
Uncemented Prostheses designed to be fixed into the bone by an initial press-fit and then bony ingrowth or
ongrowth, without using cement.
Unconfirmed prostheses construct A joint replacement which has been uploaded with either an insufficient number of elements to form
a construct, or prostheses elements which are not concordant with the procedure indicated by the
surgeon.
Unicompartmental knee Procedure where only one compartment of the knee joint is replaced, also known as partial
replacement knee replacement. The lateral (outside), medial (inside) and patellofemoral (under the knee cap)
compartments are replaced individually.
Unicondylar arthroplasty Replacement of one tibial condyle and one femoral condyle in the knee, with or without resurfacing of
the patella.
Unicondylar knee replacement See Unicondylar arthroplasty.
Unilateral operation Operation performed on one side only, e.g. left hip.
Unlinked total elbow Where the humeral and ulnar parts of a total elbow replacement are apposed but not structurally
coupled.
Summary of key facts about joint replacement during the 2020 calendar year
Hips
60% Data: average BMI
2020 2020 2020 2020 2020

54,858
NJR Patient
Consent
NJR Patient
primary average ages: 7% 88% 28.5
=
Consent
replacement
procedures
recorded on the NJR
since April 2003 66.7 69.1 Acute trauma Osteoarthritis ‘overweight’
Knees
50,904 55% Data: average BMI
NJR Patient
Consent
NJR Patient
=
Consent
replacement
procedures Unicondylar knee
recorded on the NJR
since April 2003 68.6 69.0 Osteoarthritis replacements ’obese‘
Ankles
465 41% Data: average BMI
NJR Patient
Consent
recorded on the NJR
replacement
procedures 68.8 67.1
Rheumatoid arthritis
and other inflammatory
=
since April 2010 Osteoarthritis joint problems ‘overweight’
Elbows
ent
NJR Patient
Consent
561 64% Data:
38% 45%
primary average ages: 14%
replacement
recorded on the NJR procedures 55.0 67.0
(with or without a radial head)
Radial head
replacements
Distal humeral
hemiarthroplasty
since April 2012
Shoulders
ient
NJR Patient
3,833 70% Data:
59%
Consent
primary average ages: 17% 27%
replacement
recorded on the NJR procedures 68.9 73.3 Acute trauma Osteoarthritis Elective cuff tear
since April 2012 arthropathy
For more data on clinical activity during the 2020 calendar year visit reports.njrcentre.org.uk
Information governance and patient confidentiality Terms and conditions for use of data
The NJR ensures that all patient data is processed and Do you wish to use NJR data and statistics for
handled in line with international and UK standards presentations, reports and other publications? In quoting
and within UK and European legislation: protecting and or publishing NJR data, screen shots from NJR reports
applying strict controls on the use of patient data is of the or websites we request that you reference the ‘National
highest importance. Joint Registry’. State the time-period covered, procedures
NJR data is collected via a web-based data entry included and also include reference to any other filters that
application and stored and processed in NEC Software have been applied to the data. This is particularly important
Solutions (NEC) data centre. NEC is accredited to ISO/IEC if the information is in the public domain.
27001:2013, ISO/IEC 9001:2015, ISO/IEC 20000, Cyber
Essentials Plus, and Healthcare Data Storage (HDS). NEC Where possible, include a link to www.njrcentre.org.uk so
is also registered on the NHS Data Security and Protection that the audience is able to seek out further context and
Toolkit with a status of ‘Exceeds Standards’. information on published joint replacement statistics.
For research and analysis purposes, NJR data is annually Disclaimer

linked to data from other healthcare systems using patient The National Joint Registry (NJR) produces this report
identifiers, principally a patient’s NHS number. These other using data collected, collated and provided by third parties.
datasets include the Hospital Episodes Statistics (HES) As a result of this the NJR takes no responsibility for the
service, data from the NHS England Patient Reported accuracy, currency, reliability and correctness of any data
Outcomes Measures (PROMs) programme, and Civil used or referred to in this service, nor for the accuracy,
Registration data (all provided by NHS Digital), and the currency, reliability and correctness of links or references
Patient Episode Database Wales (PEDW) (provided by to other information sources and disclaims all warranties in
NHS Wales Informatics Service). The purpose of linking relation to such data, links and references to the maximum
to these data sets is to expand and broaden the type of extent permitted by legislation.
analyses that the NJR can undertake without having to
collect additional data. This linkage has been approved by The NJR shall have no liability (including but not
the Health Research Authority under Section 251 of the limited to liability by reason of negligence) for any loss,
NHS Act 2006 on the basis of improving patient safety and damage, cost or expense incurred or arising by reason of
patient outcomes: the support provides the legal basis for any person using or relying on the data within this service
undertaking the linkage of NJR data to the health data sets and whether caused by reason of any error, omission or
listed above. misrepresentation in the presentation of data or otherwise.
Presentations of data are not to be taken as advice.
Once the datasets have been linked, patient identifiable Third parties using or relying on the data in this service
data are removed from the new dataset so that it is not do so at their own risk and will be responsible for making
possible to identify any patient. This data is then made their own assessment and should verify all relevant
available to the NJR’s statistics and analysis team at representations, statements and information with their
the University of Bristol whose processing of the data is own professional advisers.
compliant with the NHS Data Security and Protection
Toolkit. The work undertaken by the University of Bristol
is directed by the NJR’s Steering Committee and the
NJR’s Editorial Board and the results of the analyses are
published in the NJR’s Annual Report and in professional
journals. All published data is based on anonymised data, Contact:
this means that no patient could be identified.
Produced by Pad Creative Ltd www.padcreative.co.uk
NJR Service Desk

based at NEC Software Solutions UK Ltd
1st Floor, iMex Centre
575-599 Maxted Road
Hemel Hempstead
Hertfordshire
HP2 7DX
Telephone: 0845 345 9991

Fax: 0845 345 9992
Email: enquiries@njrcentre.org.uk
Website: www.njrcentre.org.uk
reports.njrcentre.org.uk
HIPS
KNEES
ANKLES
ELBOWS
SHOULDERS
At reports.njrcentre.org.uk, this document

is available to download in PDF format
along with additional data and information
on NJR progress and developments,
clinical activity as well as implant and unit-
level activity and outcomes.
Every effort has been made at the time of

publication to ensure that the information
contained in this report is accurate. If
amendments or corrections are required
after publication, they will be published on
the NJR website at www.njrcentre.org.uk
and on the dedicated NJR Reports website
at reports.njrcentre.org.uk.
/nationaljointregistry
@jointregistry

NJR 18th Annual Report 2021

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

NJR 18th Annual Report 2021

Uploaded by

Copyright:

Available Formats

HIPS

National Joint Registry

18th Annual Report

ISSN 2054-183X (Online)

NJR Editorial Board and contributors

NJR RCC Representatives

NJR Management Team

NEC Software Solutions UK Ltd

University of Bristol / University of Oxford

Pad Creative Ltd (design and production)

This work uses data provided by patients and collected by

Summary of this year’s report by the NJR

Statistics on joint replacement activity for hip,

Detailed statistical analyses on hip and knee

Indicators for hip and knee joint replacement

Information on the work of the NJR committees

Composition, attendance, declarations of

In this report and via

NJR Reports online

Top tabs: If you require

Summary of content for the NJR Annual Report 4

NJR Reports online 5

Clinical activity 2020 overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5

Navigating the NJR Reports online facility . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6

3. Outcomes after joint replacement 2003 to 2020 35

3.1 Summary of data sources, linkage and methodology 35

Information governance and patient confidentiality . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36

Patient level data linkage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39

Analytical methods and terminology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40

3.2 Outcomes after hip replacement 44

3.2.1 Overview of primary hip replacement surgery . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45

3.2.2 First revisions after primary hip surgery . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62

3.2.6 Mortality after primary hip replacement surgery . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 109

3.2.8 Overview of hip revision procedures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 116

3.2.9 Rates of hip re-revision . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 119

3.2.10 Reasons for hip re-revision . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 129

3.2.11 90-day mortality after hip revision . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 131

3.2.12 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 132

3.3 Outcomes after knee replacement 135

3.3.1 Overview of primary knee replacement surgery . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 136

3.3.2 First revision after primary knee surgery . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 148

3.3.5 Mortality after primary knee surgery . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 191

3.3.6 Overview of knee revisions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 193

3.3.7 Rates of knee re-revision . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 196

3.3.8 Reason for knee re-revision . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 208

3.3.9 90-day mortality after knee revision . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 210

3.3.10 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 210

3.4 Outcomes after ankle replacement 213

3.4.1 Overview of primary ankle replacement surgery . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 214

3.4.2 Revisions after primary ankle surgery . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 221

3.4.3 Mortality after primary ankle replacement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 226

3.4.4 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 228

3.5 Outcomes after elbow replacement 229

3.5.1 Overview of primary elbow replacement surgery . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 230

3.5.2 Revisions after primary elbow replacement surgery . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 242

3.5.3 Mortality after primary elbow replacement surgery . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 252

3.5.4 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 256

3.6.1 Overview of primary shoulder replacement surgery . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 258

3.6.2 Revisions after primary shoulder replacement surgery . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 276

3.6.4 Mortality after primary shoulder replacement surgery . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 298

3.6.5 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 303

3.7 In-depth studies 305

Published papers 2020-2021 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 330

4. Implant and unit-level activity and outcomes 333